A high-level knowledge-representation framework that can be used to solve problems declaratively based onabductive reasoning. It extends normallogic programming by allowing some predicates to be incompletely defined, declared as abducible predicates.
A form oflogical inference which starts with an observation or set of observations then seeks to find the simplest and most likely explanation. This process, unlikedeductive reasoning, yields a plausible conclusion but does notpositively verify it.[1] abductive inference,[1] or retroduction[2]
The removal of a component of an AI system. Anablation study aims to determine the contribution of a component to an AI system by removing the component, and then analyzing the resultant performance of the system.[3]
Amathematical model fordata types, where a data type is defined by its behavior (semantics) from the point of view of auser of the data, specifically in terms of possible values, possible operations on data of this type, and the behavior of these operations.
The process of removing physical, spatial, or temporal details[4] orattributes in the study of objects orsystems in order to more closely attend to other details of interest[5]
A perceived increase in the rate oftechnological change throughout history, which may suggest faster and more profound change in the future and may or may not be accompanied by equally profound social and cultural change.
An area ofmachine learning concerned with creation and modification of software agent's knowledge about effects and preconditions of the actions that can be executed within its environment. This knowledge is usually represented in logic-based action description language and used as the input for automated planners.
A way of characterizing the most basic problem of intelligent systems: what to do next. In artificial intelligence and computational cognitive science, "the action selection problem" is typically associated with intelligent agents and animats—artificial systems that exhibit complex behaviour in an agent environment.
A kind ofartificial neural network that is based on Takagi–Sugeno fuzzyinference system. The technique was developed in the early 1990s.[7][8] Since it integrates both neural networks andfuzzy logic principles, it has potential to capture the benefits of both in a singleframework. Its inference system corresponds to a set of fuzzyIF–THEN rules that have learning capability to approximate nonlinear functions.[9] Hence, ANFIS is considered to be a universal estimator.[10] For using the ANFIS in a more efficient and optimal way, one can use the best parameters obtained by genetic algorithm.[11][12]
Incomputer science, specifically inalgorithms related topathfinding, aheuristic function is said to be admissible if it never overestimates the cost of reaching the goal, i.e. the cost it estimates to reach the goal is not higher than the lowest possible cost from the current point in the path.[13]
The study and development of systems and devices that can recognize, interpret, process, and simulate humanaffects. Affective computing is an interdisciplinary field spanningcomputer science,psychology, andcognitive science.[14][15]
In the field of artificial intelligence, the most difficult problems are informally known as AI-complete or AI-hard, implying that the difficulty of these computational problems is equivalent to that of solving the central artificial intelligence problem—making computers as intelligent as people, orstrong AI.[19] To call a problem AI-complete reflects an attitude that it would not be solved by a simple specific algorithm.
A property of analgorithm which relates to the number ofcomputational resources used by the algorithm. An algorithm must beanalyzed to determine its resource usage, and the efficiency of an algorithm can be measured based on usage of different resources. Algorithmic efficiency can be thought of as analogous to engineeringproductivity for a repeating or continuous process.
The determination of thecomputational complexity of algorithms, that is the amount of time, storage and/or other resources necessary toexecute them. Usually, this involves determining afunction that relates the length of an algorithm's input to the number of steps it takes (itstime complexity) or the number of storage locations it uses (itsspace complexity).
A form ofdeclarative programming oriented towards difficult (primarilyNP-hard)search problems. It is based on thestable model (answer set) semantics oflogic programming. In ASP, search problems are reduced to computing stable models, andanswer set solvers—programs for generating stable models—are used to perform search.
A set of subroutine definitions,communication protocols, and tools for building software. In general terms, it is a set of clearly defined methods of communication among various components. A good API makes it easier to develop acomputer program by providing all the building blocks, which are then put together by theprogrammer. An API may be for a web-based system,operating system,database system, computer hardware, orsoftware library.
The technique of findingstrings that match apattern approximately (rather than exactly). The problem of approximate string matching is typically divided into two sub-problems: finding approximatesubstring matches inside a given string and finding dictionary strings that match the pattern approximately.
A way to deal with contentious information and draw conclusions from it. In an abstract argumentation framework,[25] entry-level information is a set of abstract arguments that, for instance, represent data or a proposition. Conflicts between arguments are represented by abinary relation on the set of arguments. In concrete terms, you represent an argumentation framework with adirected graph such that the nodes are the arguments, and the arrows represent the attack relation. There exist some extensions of the Dung's framework, like the logic-based argumentation frameworks[26] or the value-based argumentation frameworks.[27]
Anyintelligence demonstrated bymachines, in contrast to the natural intelligence displayed by humans and other animals. Incomputer science, AI research is defined as the study of "intelligent agents": any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.[28] Colloquially, the term "artificial intelligence" is applied when a machine mimics "cognitive" functions that humans associate with otherhuman minds, such as "learning" and "problem solving".[29]
An international, nonprofit, scientific society devoted to promote research in, and responsible use of,artificial intelligence. AAAI also aims to increase public understanding of artificial intelligence (AI), improve the teaching and training of AI practitioners, and provide guidance for research planners and funders concerning the importance and potential of current AI developments and future directions.[30]
Machine learning-basedattention is a mechanism mimickingcognitive attention. It calculates "soft" weights for each word, more precisely for its embedding, in thecontext window. It can do it either in parallel (such as intransformers) or sequentially (such as inrecursive neural networks). "Soft" weights can change during each runtime, in contrast to "hard" weights, which are (pre-)trained and fine-tuned and remain frozen afterwards. Multiple attention heads are used in transformer-basedlarge language models.
An interactive experience of a real-world environment where the objects that reside in the real-world are "augmented" by computer-generated perceptual information, sometimes across multiple sensory modalities, includingvisual,auditory,haptic,somatosensory, andolfactory.[31]
Theself-managing characteristics ofdistributed computing resources, adapting to unpredictable changes while hiding intrinsic complexity to operators and users. Initiated byIBM in 2001, this initiative ultimately aimed to develop computer systems capable of self-management, to overcome the rapidly growing complexity of computingsystems management, and to reduce the barrier that complexity poses to further growth.[33]
A method used inartificial neural networks to calculate a gradient that is needed in the calculation of theweights to be used in the network.[38] Backpropagation is shorthand for "the backward propagation of errors", since an error is computed at the output and distributed backwards throughout the network's layers. It is commonly used to traindeep neural networks,[39] a term referring to neural networks with more than onehidden layer.[40]
In computer vision, the bag-of-words model (BoW model) can be applied toimage classification, by treatingimage features as words. In document classification, abag of words is asparse vector of occurrence counts of words; that is, a sparsehistogram over the vocabulary. Incomputer vision, abag of visual words is a vector of occurrence counts of a vocabulary of local image features.
A technique for improving the performance and stability ofartificial neural networks. It is a technique to provide any layer in a neural network with inputs that are zero mean/unit variance.[48] Batch normalization was introduced in a 2015 paper.[49][50] It is used to normalize the input layer by adjusting and scaling the activations.
A formalism and a methodology for having a technique to specifyprobabilistic models and solve problems when less than the necessary information is available.
A population-basedsearch algorithm which was developed by Pham, Ghanbarzadeh and et al. in 2005.[51] It mimics the food foraging behaviour of honey bee colonies. In its basic version the algorithm performs a kind of neighborhood search combined with global search, and can be used for bothcombinatorial optimization andcontinuous optimization. The only condition for the application of the bees algorithm is that some measure of distance between the solutions is defined. The effectiveness and specific abilities of the bees algorithm have been proven in a number of studies.[52][53][54][55]
Amathematical model ofplan execution used incomputer science,robotics,control systems andvideo games. They describe switchings between a finite set of tasks in a modular fashion. Their strength comes from their ability to create very complex tasks composed of simple tasks, without worrying how the simple tasks are implemented. BTs present some similarities tohierarchical state machines with the key difference that the main building block of a behavior is a task rather than a state. Its ease of human understanding make BTs less error-prone and very popular in the game developer community. BTs have shown to generalize several other control architectures.[57][58]
A software model developed for programmingintelligent agents. Superficially characterized by the implementation of an agent'sbeliefs,desires andintentions, it actually uses these concepts to solve a particular problem in agent programming. In essence, it provides a mechanism for separating the activity of selecting a plan (from a plan library or an external planner application) from the execution of currently active plans. Consequently, BDI agents are able to balance the time spent on deliberating about plans (choosing what to do) and executing those plans (doing it). A third activity, creating the plans in the first place (planning), is not within the scope of the model, and is left to the system designer and programmer.
A mathematical notation that describes thelimiting behavior of afunction when theargument tends towards a particular value or infinity. It is a member of a family of notations invented byPaul Bachmann,[60]Edmund Landau,[61] and others, collectively called Bachmann–Landau notation or asymptotic notation.
Atreedata structure in which each node has at most twochildren, which are referred to as theleft child and theright child. Arecursive definition using justset theory notions is that a (non-empty) binary tree is atuple (L,S,R), whereL andR are binary trees or theempty set andS is asingleton set.[62] Some authors allow the binary tree to be the empty set as well.[63]
Anartificial intelligence approach based on theblackboard architectural model,[64][65][66][67] where a common knowledge base, the "blackboard", is iteratively updated by a diverse group of specialist knowledge sources, starting with a problem specification and ending with a solution. Each knowledge source updates the blackboard with a partial solution when its internal constraints match the blackboard state. In this way, the specialists work together to solve the problem.
The problem of determining if there exists aninterpretation thatsatisfies a givenBooleanformula. In other words, it asks whether the variables of a given Boolean formula can be consistently replaced by the values TRUE or FALSE in such a way that the formula evaluates to TRUE. If this is the case, the formula is calledsatisfiable. On the other hand, if no such assignment exists, the function expressed by the formula isFALSE for all possible variable assignments and the formula isunsatisfiable. For example, the formula "a AND NOTb" is satisfiable because one can find the valuesa = TRUE andb = FALSE, which make (a AND NOTb) = TRUE. In contrast, "a AND NOTa" is unsatisfiable.
A technology that employs the latest findings inneuroscience. The term was first introduced by the Artificial Intelligence Laboratory inZurich, Switzerland, in the context of theROBOY project.[69] Brain Technology can be employed in robots,[70]know-how management systems[71] and any other application with self-learning capabilities. In particular, Brain Technology applications allow the visualization of the underlying learning architecture often coined as "know-how maps".
A very generalproblem-solving technique andalgorithmic paradigm that consists of systematically enumerating all possible candidates for the solution and checking whether each candidate satisfies the problem's statement.
Amachine learning system that is a type ofartificial neural network (ANN) that can be used to better model hierarchical relationships. The approach is an attempt to more closely mimic biological neural organization.[72]
A field ofrobotics that attempts to invoke cloud technologies such ascloud computing,cloud storage, and otherInternet technologies centred on the benefits of converged infrastructure and shared services for robotics. When connected to the cloud, robots can benefit from the powerful computation, storage, and communication resources of moderndata center in the cloud, which can process and share information from various robots or agent (other machines, smart objects, humans, etc.). Humans can also delegate tasks to robots remotely throughnetworks. Cloud computing technologies enable robot systems to be endowed with powerful capability whilst reducing costs through cloud technologies. Thus, it is possible to build lightweight, low cost, smarter robots have intelligent "brain" in the cloud. The "brain" consists ofdata center,knowledge base, task planners,deep learning, information processing, environment models, communication support, etc.[74][75][76][77]
An incremental system for hierarchicalconceptual clustering. COBWEB was invented by ProfessorDouglas H. Fisher, currently at Vanderbilt University.[78][79] COBWEB incrementally organizes observations into aclassification tree. Each node in a classification tree represents a class (concept) and is labeled by a probabilistic concept that summarizes the attribute-value distributions of objects classified under the node. This classification tree can be used to predict missing attributes or the class of a new object.[80]
TheInstitute of Creative Technologies defines cognitive architecture as: "hypothesis about the fixed structures that provide a mind, whether in natural or artificial systems, and how they work together – in conjunction with knowledge and skills embodied within the architecture – to yield intelligent behavior in a diversity of complex environments."[81]
In general, the term cognitive computing has been used to refer to new hardware and/or software thatmimics the functioning of thehuman brain[82][83][84][85][86][87] and helps to improve human decision-making.[88] In this sense, CC is a new type of computing with the goal of more accurate models of how the human brain/mind senses,reasons, and responds to stimulus.
A type ofartificial neural network using adivide and conquer strategy in which the responses of multiple neural networks (experts) are combined into a single response.[91] The combined response of the committee machine is supposed to be superior to those of its constituent experts. Compareensembles of classifiers.
Inartificial intelligence research, commonsense knowledge consists of facts about the everyday world, such as "Lemons are sour", that all humans are expected to know. The first AI program to address common sense knowledge wasAdvice Taker in 1959 by John McCarthy.[92]
A branch of artificial intelligence concerned with simulating the human ability to make presumptions about the type and essence of ordinary situations they encounter every day.[93]
Focuses on classifying computational problems according to their inherent difficulty, and relating these classes to each other. A computational problem is a task solved by a computer. A computation problem is solvable by mechanical application of mathematical steps, such as an algorithm.
Aninterdisciplinary field concerned with the statistical or rule-based modeling ofnatural language from a computational perspective, as well as the study of appropriate computational approaches to linguistic questions.
The theory, experimentation, and engineering that form the basis for the design and use ofcomputers. It involves the study ofalgorithms that process, store, and communicatedigitalinformation. Acomputer scientist specializes in the theory of computation and the design of computational systems.[115]
Inpredictive analytics andmachine learning, the concept drift means that the statistical properties of the target variable, which the model is trying to predict, change over time in unforeseen ways. This causes problems because the predictions become less accurate as time passes.
In the study ofpath-finding problems inartificial intelligence, aheuristic function is said to be consistent, or monotone, if its estimate is always less than or equal to the estimated distance from any neighboring vertex to the goal, plus the cost of reaching that neighbor.
Amachine learning and inference framework that augments the learning of conditional (probabilistic or discriminative) models with declarative constraints.
A form ofconstraint programming, in whichlogic programming is extended to include concepts fromconstraint satisfaction. A constraint logic program is a logic program that contains constraints in the body of clauses. An example of a clause including a constraint isA(X,Y):-X+Y>0,B(X),C(Y). In this clause,X+Y>0 is a constraint;A(X,Y),B(X), andC(Y) areliterals as in regular logic programming. This clause states one condition under which the statementA(X,Y) holds:X+Y is greater than zero and bothB(X) andC(Y) are true.
A language whosephonology,grammar, andvocabulary are consciously devised, instead of having developednaturally. Constructed languages may also be referred to as artificial, planned, or invented languages.[120]
Incontrol systems engineering is a subfield of mathematics that deals with the control of continuously operatingdynamical systems in engineered processes and machines. The objective is to develop a control model for controlling such systems using a control action in an optimum manner withoutdelay or overshoot and ensuring controlstability.
Indeep learning, a convolutional neural network (CNN, or ConvNet) is a class of deepneural network most commonly applied to image analysis. CNNs use a variation ofmultilayer perceptrons designed to require minimalpreprocessing.[121] They are also known as shift invariant or space invariant artificial neural networks (SIANN), based on their shared-weights architecture andtranslation invariance characteristics.[122][123]
The Dartmouth Summer Research Project on Artificial Intelligence was the name of a 1956 summer workshop now considered by many[128][129] (though not all[130]) to be theseminal event forartificial intelligence as a field.
The process of integrating multiple data sources to produce more consistent, accurate, and useful information than that provided by any individual data source.[131]
The process of combiningdata residing in different sources and providing users with a unified view of them.[132] This process becomes significant in a variety of situations, which include both commercial (such as when two similar companies need to merge theirdatabases) and scientific (combining research results from differentbioinformatics repositories, for example) domains. Data integration appears with increasing frequency as the volume (that is,big data) and the need to share existing dataexplodes.[133] It has become the focus of extensive theoretical work, and numerous open problems remain unsolved.
An interdisciplinary field that uses scientific methods, processes, algorithms and systems to extractknowledge and insights fromdata in various forms, both structured and unstructured,[134][135] similar todata mining. Data science is a "concept to unify statistics, data analysis,machine learning, and their related methods" in order to "understand and analyze actual phenomena" with data.[136] It employs techniques and theories drawn from many fields within the context ofmathematics,statistics,information science, andcomputer science.
A collection ofdata. Most commonly a data set corresponds to the contents of a singledatabase table, or a single statisticaldata matrix, where everycolumn of the table represents a particular variable, and eachrow corresponds to a given member of the data set in question. The data set lists values for each of the variables, such as height and weight of an object, for each member of the data set. Each value is known as a datum. The data set may comprise data for one or more members, corresponding to the number of rows.
A system used forreporting anddata analysis.[137] DWs are central repositories of integrated data from one or more disparate sources. They store current and historical data in one single place[138]
Aninformation system that supports business or organizationaldecision-making activities. DSSs serve the management, operations and planning levels of an organization (usually mid and higher management) and help people make decisions about problems that may be rapidly changing and not easily specified in advance—i.e. unstructured and semi-structured decision problems. Decision support systems can be either fully computerized or human-powered, or a combination of both.
The study of the reasoning underlying anagent's choices.[140] Decision theory can be broken into two branches:normative decision theory, which gives advice on how to make thebest decisions given a set of uncertain beliefs and a set ofvalues, and descriptive decision theory which analyzes how existing, possibly irrational agents actually make decisions.
Uses adecision tree (as apredictive model) to go from observations about an item (represented in the branches) to conclusions about the item's target value (represented in the leaves). It is one of the predictive modeling approaches used instatistics,data mining andmachine learning.
was achess-playing computer developed byIBM. It is known for being the first computer chess-playing system to win both a chess game and a chess match against a reigning world champion under regular time controls.
A family of formalknowledge representation languages. Many DLs are more expressive thanpropositional logic but less expressive thanfirst-order logic. In contrast to the latter, the core reasoning problems for DLs are (usually)decidable, and efficient decision procedures have been designed and implemented for these problems. There are general, spatial, temporal, spatiotemporal, and fuzzy descriptions logics, and each description logic features a different balance between DL expressivity andreasoningcomplexity by supporting different sets of mathematical constructors.[151]
A scientific field which aims at studying the developmental mechanisms, architectures, and constraints that allow lifelong and open-ended learning of new skills and new knowledge in embodiedmachines.
Concerned with the development of algorithms and techniques that are able to determine whether the behaviour of a system is correct. If the system is not functioning correctly, the algorithm should be able to determine, as accurately as possible, which part of the system is failing, and which kind of fault it is facing. The computation is based onobservations, which provide information on the current behaviour.
A computer system intended to converse with a human with a coherent structure. Dialogue systems have employed text, speech, graphics, haptics, gestures, and other modes for communication on both the input and output channel.
Inmachine learning,diffusion models, also known asdiffusion probabilistic models orscore-based generative models, are a class oflatent variable models. They areMarkov chains trained usingvariational inference.[152] The goal of diffusion models is to learn the latent structure of a dataset by modeling the way in which data points diffuse through thelatent space. Incomputer vision, this means that a neural network is trained todenoise images blurred withGaussian noise by learning to reverse the diffusion process.[153][154] It mainly consists of three major components: the forward process, the reverse process, and the sampling procedure.[155] Three examples of generic diffusion modeling frameworks used in computer vision are denoising diffusion probabilistic models, noise conditioned score networks, and stochastic differential equations.[156]
The process of reducing the number of random variables under consideration[157] by obtaining a set of principal variables. It can be divided intofeature selection andfeature extraction.[158]
Any system with a countable number of states. Discrete systems may be contrasted with continuous systems, which may also be called analog systems. A final discrete system is often modeled with a directedgraph and is analyzed for correctness and complexity according tocomputational theory. Because discrete systems have a countable number of states, they may be described in precisemathematical models. Acomputer is afinite-state machine that may be viewed as a discrete system. Because computers are often used to model not only other discrete systems but continuous systems as well, methods have been developed to represent real-world continuous systems as discrete systems. One such method involves sampling a continuous signal atdiscrete time intervals.
A subfield ofartificial intelligence research dedicated to the development of distributed solutions for problems. DAI is closely related to and a predecessor of the field ofmulti-agent systems.[159]
A phenomenon instatistics andmachine learning where amodel with a small number ofparameters and a model with an extremely large number of parameters have a smalltest error, but a model whose number of parameters is about the same as the number ofdata points used to train the model will have a large error.[160] This phenomenon has been considered surprising, as it contradicts assumptions aboutoverfitting in classical machine learning.[161]
A logical framework dealing with knowledge and information change. Typically, DEL focuses on situations involving multipleagents and studies how their knowledge changes whenevents occur.
A learning method in which the system tries to construct a general, input-independent target function during training of the system, as opposed tolazy learning, wheregeneralization beyond the training data is delayed until a query is made to the system.[162]
Arecurrent neural network with a sparsely connectedhidden layer (with typically 1% connectivity). The connectivity and weights of hidden neurons are fixed and randomly assigned. The weights of output neurons can be learned so that the network can (re)produce specific temporal patterns. The main interest of this network is that although its behaviour is non-linear, the only weights that are modified during training are for the synapses that connect the hidden neurons to output neurons. Thus, the error function is quadratic with respect to the parameter vector and can be differentiated easily to a linear system.[167][168]
Anintelligent agent that interacts with the environment through a physical body within that environment. Agents that are represented graphically with a body, for example a human or a cartoon animal, are also called embodied agents, although they have only virtual, not physical, embodiment.[169]
An interdisciplinary field of research, the aim of which is to explain the mechanisms underlying intelligent behavior. It comprises three main methodologies: 1) the modeling of psychological and biological systems in a holistic manner that considers the mind and body as a single entity, 2) the formation of a common set of general principles of intelligent behavior, and 3) the experimental use of robotic agents in controlled environments.
Inmachine learning, particularly in the creation ofartificial neural networks, an epoch is training the model for one cycle through the full training dataset. Small models are typically trained for as many epochs as it takes to reach the best performance on the validation dataset. The largest models may train for only one epoch.
A computer system that emulates the decision-making ability of a human expert.[177] Expert systems are designed to solve complex problems byreasoning through bodies of knowledge, represented mainly asif–then rules rather than through conventionalprocedural code.[178]
A type ofclassification tree. Fast-and-frugal trees can be used as decision-making tools which operate as lexicographic classifiers, and, if required, associate an action (decision) to each class or category.[179]
An individual measurable property or characteristic of a phenomenon.[180] Incomputer vision andimage processing, a feature is a piece of information about the content of an image; typically about whether a certain region of the image has certain properties. Features may be specific structures in an image (such as points, edges, or objects), or the result of a generalneighborhood operation or feature detection applied to the image.
Inmachine learning,pattern recognition, andimage processing, feature extraction starts from an initial set of measured data and builds derived values (features) intended to be informative and non-redundant, facilitating the subsequent learning andgeneralization steps, and in some cases leading to better human interpretations.
Inmachine learning,feature learning or representation learning[181] is a set of techniques that allows a system to automatically discover the representations needed for feature detection orclassification from raw data. This replaces manualfeature engineering and allows a machine to both learn the features and use them to perform a specific task.
Inmachine learning andstatistics, feature selection, also known as variable selection, attribute selection or variable subset selection, is the process of selecting a subset of relevantfeatures (variables, predictors) for use in model construction.
Amachine learning technique that allows for training models on multiple devices with decentralized data, thus helping preserve the privacy of individual users and their data.
A collection offormal systems used inmathematics,philosophy,linguistics, andcomputer science. First-order logic usesquantified variables over non-logical objects and allows the use of sentences that contain variables, so that rather than propositions such asSocrates is a man one can have expressions in the form "there exists X such that X isSocrates and X is a man" andthere exists is a quantifier whileX is a variable.[182] This distinguishes it frompropositional logic, which does not use quantifiers or relations.[183]
A condition that can change over time. In logical approaches to reasoning about actions, fluents can be represented infirst-order logic bypredicates having an argument that depends on time.
One of the two main methods of reasoning when using aninference engine and can be describedlogically as repeated application ofmodus ponens. Forward chaining is a popular implementation strategy forexpert systems,businesses andproduction rule systems. The opposite of forward chaining isbackward chaining. Forward chaining starts with the availabledata and uses inference rules to extract more data (from an end user, for example) until agoal is reached. Aninference engine using forward chaining searches the inference rules until it finds one where theantecedent (If clause) is known to be true. When such a rule is found, the engine can conclude, or infer, theconsequent (Then clause), resulting in the addition of newinformation to its data.[184]
An artificial intelligencedata structure used to divideknowledge into substructures by representing "stereotyped situations". Frames are the primary data structure used in artificial intelligenceframe language.
A technology used forknowledge representation in artificial intelligence. Frames are stored asontologies ofsets and subsets of theframe concepts. They are similar to class hierarchies inobject-oriented languages although their fundamental design goals are different. Frames are focused on explicit and intuitive representation of knowledge whereas objects focus onencapsulation andinformation hiding. Frames originated in AI research and objects primarily insoftware engineering. However, in practice the techniques and capabilities of frame and object-oriented languages overlap significantly.
A hypotheticalartificial general intelligence (AGI) that would have a positive effect on humanity. It is a part of theethics of artificial intelligence and is closely related tomachine ethics. While machine ethics is concerned with how an artificially intelligent agent should behave, friendly artificial intelligence research is focused on how to practically bring about this behaviour and ensuring it is adequately constrained.
Acontrol system based onfuzzy logic—amathematical system that analyzesanalog input values in terms oflogical variables that take on continuous values between 0 and 1, in contrast to classical ordigital logic, which operates on discrete values of either 1 or 0 (true or false, respectively).[187][188]
A simple form for themany-valued logic, in which thetruth values of variables may have any degree of "Truthfulness" that can be represented by any real number in the range between 0 (as in Completely False) and 1 (as in Completely True) inclusive. Consequently, It is employed to handle the concept of partial truth, where the truth value may range between completely true and completely false. In contrast toBoolean logic, where the truth values of variables may have the integer values 0 or 1 only.
In classicalset theory, the membership of elements in a set is assessed in binary terms according to abivalent condition — an element either belongs or does not belong to the set. By contrast, fuzzy set theory permits the gradual assessment of the membership of elements in a set; this is described with the aid of amembership function valued in the real unit interval [0, 1]. Fuzzy sets generalize classical sets, since theindicator functions (akacharacteristic functions) of classical sets are special cases of the membership functions of fuzzy sets, if the latter only take values 0 or 1.[189] In fuzzy set theory, classical bivalent sets are usually calledcrisp sets. The fuzzy set theory can be used in a wide range of domains in which information is incomplete or imprecise, such asbioinformatics.[190]
The concept that humans, other animals, andartificial neural networks use past learning in present situations of learning if the conditions in the situations are regarded as similar.[195]
Alarge language model based on thetransformer architecture that generates text. It is first pretrained to predict the nexttoken in texts (a token is typically a word, subword, or punctuation). After their pretraining, GPT models can generate human-like text by repeatedly predicting the token that they would expect to follow. GPT models are usually also fine-tuned, for example withreinforcement learning from human feedback to reducehallucination or harmful behaviour, or to format the output in a conversationnal format.[202]
Anoperator used ingenetic algorithms to guide the algorithm towards a solution to a given problem. There are three main types of operators (mutation,crossover andselection), which must work in conjunction with one another in order for the algorithm to be successful.
Amachine learning technique based onboosting in a functional space, where the target ispseudo-residuals instead ofresiduals as in traditional boosting.
In mathematics, and more specifically ingraph theory, a graph is a structure amounting to a set of objects in which some pairs of the objects are in some sense "related". The objects correspond to mathematical abstractions calledvertices (also callednodes orpoints) and each of the related pairs of vertices is called anedge (also called anarc orline).[204]
Adatabase that usesgraph structures forsemantic queries withnodes,edges, and properties to represent and store data. A key concept of the system is thegraph (oredge orrelationship), which directly relates data items in the store a collection of nodes of data and edges representing the relationships between the nodes. The relationships allow data in the store to be linked together directly, and in many cases retrieved with one operation. Graph databases hold the relationships between data as a priority. Querying relationships within a graph database is fast because they are perpetually stored within the database itself. Relationships can be intuitively visualized using graph databases, making it useful for heavily inter-connected data.[205][206]
The process of visiting (checking and/or updating) each vertex in agraph. Such traversals are classified by the order in which the vertices are visited.Tree traversal is a special case of graph traversal.
A technique designed forsolving a problem more quickly when classic methods are too slow, or for finding an approximate solution when classic methods fail to find any exact solution. This is achieved by trading optimality, completeness,accuracy, orprecision for speed. In a way, it can be considered a shortcut. A heuristic function, also called simply a heuristic, is afunction that ranks alternatives insearch algorithms at each branching step based on available information to decide which branch to follow. For example, it may approximate the exact solution.[207]
Aheuristic search method that seeks to automate the process of selecting, combining, generating, or adapting several simpler heuristics (or components of such heuristics) to efficiently solve computational search problems, often by the incorporation ofmachine learning techniques. One of the motivations for studying hyper-heuristics is to build systems which can handle classes of problems rather than solving just one problem.[208][209][210]
A decision boundary inmachine learningclassifiers that partitions the input space into two or more sections, with each section corresponding to a unique class label.
A method ofmachine learning, in which input data is continuously used to extend the existing model's knowledge i.e. to further train the model. It represents a dynamic technique ofsupervised andunsupervised learning that can be applied when training data becomes available gradually over time or its size is out of system memory limits. Algorithms that can facilitate incremental learning are known as incremental machine learning algorithms.
The merging of information from heterogeneous sources with differing conceptual, contextual and typographical representations. It is used indata mining and consolidation of data from unstructured or semi-structured resources. Typically,information integration refers to textual representations of knowledge but is sometimes applied torich-media content. Information fusion, which is a related term, involves the combination of information into a new set of information towards reducing redundancy and uncertainty.[131]
A possible outcome of humanity buildingartificial general intelligence (AGI). AGI would be capable of recursive self-improvement leading to rapid emergence of ASI (artificial superintelligence), the limits of which are unknown, at the time of the technological singularity.
Anautonomous entity which acts, directing its activity towards achieving goals (i.e. it is anagent), upon anenvironment using observation through sensors and consequent actuators (i.e. it is intelligent). Intelligent agents may alsolearn or useknowledge to achieve their goals. They may be very simple orvery complex.
Alsovirtual assistant orpersonal digital assistant.
Asoftware agent that can perform tasks or services for an individual based on verbal commands. Sometimes the term "chatbot" is used to refer to virtual assistants generally or specifically accessed byonline chat (or in some cases online chat programs that are exclusively for entertainment purposes). Some virtual assistants are able to interpret human speech and respond via synthesized voices. Users can ask their assistants questions, controlhome automation devices and media playback via voice, and manage other basic tasks such as email, to-do lists, and calendars with verbal commands.[213]
Anintelligent agent is intrinsically motivated to act if the information content alone, of the experience resulting from the action, is the motivating factor. Information content in this context is measured in theinformation theory sense as quantifying uncertainty. A typical intrinsic motivation is to search for unusual (surprising) situations, in contrast to a typical extrinsic motivation such as the search for food. Intrinsically motivated artificial agents display behaviours akin toexploration andcuriosity.[214]
A graphical breakdown of a question that dissects it into its different components vertically and that progresses into details as it reads to the right.[215]: 47 Issue trees are useful inproblem solving to identify the root causes of a problem as well as to identify its potential solutions. They also provide a reference point to see how each piece fits into the whole picture of a problem.[216]
A well-knownknowledge representation system in the tradition ofsemantic networks andframes; that is, it is aframe language. The system is an attempt to overcome semantic indistinctness in semantic network representations and to explicitly represent conceptual information as a structured inheritance network.[218][219][220]
The process used to define the rules and ontologies required for aknowledge-based system. The phrase was first used in conjunction withexpert systems to describe the initial tasks associated with developing an expert system, namely finding and interviewingdomain experts and capturing their knowledge viarules,objects, andframe-basedontologies.
Acomputer program thatreasons and uses aknowledge base tosolvecomplex problems. The term is broad and refers to many different kinds of systems. The one common theme that unites all knowledge based systems is an attempt to represent knowledge explicitly and areasoning system that allows it to derive new knowledge. Thus, a knowledge-based system has two distinguishing features: aknowledge base and aninference engine.
The creation ofknowledge from structured (relational databases,XML) and unstructured (text, documents,images) sources. The resulting knowledge needs to be in a machine-readable and machine-interpretable format and mustrepresent knowledge in a manner that facilitates inferencing. Although it is methodically similar toinformation extraction andETL, the main criterion is that the extraction result goes beyond the creation of structured information or the transformation into arelational schema. It requires either the reuse of existingformal knowledge (reusing identifiers orontologies) or the generation of a schema based on the source data.
A computer language designed to enable systems to share and reuse information fromknowledge-based systems. KIF is similar toframe languages such asKL-ONE andLOOM but unlike such language its primary role is not intended as a framework for the expression or use of knowledge but rather for the interchange of knowledge between systems. The designers of KIF likened it toPostScript. PostScript was not designed primarily as a language to store and manipulate documents but rather as an interchange format for systems and devices to share documents. In the same way KIF is meant to facilitate sharing of knowledge across different systems that use different languages, formalisms, platforms, etc.
A method ofvector quantization, originally fromsignal processing, that aims topartitionn observations intokclusters in which each observation belongs to the cluster with the nearestmean (cluster centers or clustercentroid), serving as a prototype of the cluster.
Alanguage model with a large number ofparameters (typically at least a billion) that are adjusted during training. Due to its size, it requires a lot of data and computing capability to train. Large language models are usually based on thetransformer architecture.[224]
Inmachine learning, lazy learning is a learning method in whichgeneralization of thetraining data is, in theory, delayed until a query is made to the system, as opposed to ineager learning, where the system tries to generalize the training data before receiving queries.
An artificialrecurrent neural network architecture[226] used in the field ofdeep learning. Unlike standardfeedforward neural networks, LSTM has feedback connections that make it a "general purpose computer" (that is, it can compute anything that aTuring machine can).[227] It can not only process single data points (such as images), but also entire sequences of data (such as speech or video).
LoRA stands for Low-Rank Adaptation. It is a method used to fine-tune large models by updating only a small, targeted part of the model. This makes it quicker and less resource-intensive to adapt the model to specific tasks or new datasets.
The technology and methods used to provide imaging-based automatic inspection and analysis for such applications as automatic inspection,process control, and robot guidance, usually in industry. Machine vision is a term encompassing a large number of technologies, software and hardware products, integrated systems, actions, methods and expertise. Machine vision as asystems engineering discipline can be considered distinct fromcomputer vision, a form ofcomputer science. It attempts to integrate existing technologies in new ways and apply them to solve real world problems. The term is the prevalent one for these functions in industrial automation environments but is also used for these functions in other environments such as security and vehicle guidance.
Astochastic model describing asequence of possible events in which the probability of each event depends only on the state attained in the previous event.[228][229]
The capability of a computer system to interpret data in a manner that is similar to the way humans use their senses to relate to the world around them.[233][234][235]
A field ineconomics andgame theory that takes anengineering approach to designing economic mechanisms orincentives, toward desired objectives, instrategic settings, where players actrationally. Because it starts at the end of the game, then goes backwards, it is also called reverse game theory. It has broad applications, from economics and politics (markets, auctions, voting procedures) to networked-systems (internet interdomain routing, sponsored search auctions).
Allows for an in-depth insight into the molecular mechanisms of a particular organism. In particular, these models correlate thegenome with molecular physiology.[238]
Incomputer science andmathematical optimization, a metaheuristic is a higher-levelprocedure orheuristic designed to find, generate, or select a heuristic (partialsearch algorithm) that may provide a sufficiently good solution to anoptimization problem, especially with incomplete or imperfect information or limited computation capacity.[239][240] Metaheuristics sample a set of solutions which is too large to be completely sampled.
Incomputer science, model checking or property checking is, for a given model of a system, exhaustively and automatically checking whether this model meets a givenspecification. Typically, one has hardware or software systems in mind, whereas the specification contains safety requirements such as the absence ofdeadlocks and similar critical states that can cause the system tocrash. Model checking is a technique for automatically verifying correctness properties offinite-state systems.
Inpropositional logic,modus tollens is avalidargument form and arule of inference. It is an application of the general truth that if a statement is true, then so is itscontrapositive. The inference rulemodus tollens asserts that theinference fromP implies Q tothe negation of Q implies the negation of P is valid.
A variant ofparticle swarm optimization (PSO) based on the use of multiple sub-swarms instead of one (standard) swarm. The general approach in multi-swarm optimization is that each sub-swarm focuses on a specific region while a specific diversification method decides where and when to launch the sub-swarms. The multi-swarm framework is especially fitted for the optimization on multi-modal problems, where multiple (local) optima exist.
Agenetic operator used to maintaingenetic diversity from one generation of a population ofgenetic algorithmchromosomes to the next. It is analogous to biologicalmutation. Mutation alters one or more gene values in a chromosome from its initial state. In mutation, the solution may change entirely from the previous solution. Hence GA can come to a better solution by using mutation. Mutation occurs during evolution according to a user-definable mutation probability. This probability should be set low. If it is set too high, the search will turn into a primitive random search.
An earlybackward chainingexpert system that usedartificial intelligence to identify bacteria causing severe infections, such asbacteremia andmeningitis, and to recommendantibiotics, with the dosage adjusted for patient's body weight – the name derived from the antibiotics themselves, as many antibiotics have the suffix "-mycin". The MYCIN system was also used for the diagnosis of blood clotting diseases.
An approach used in computer science forrepresenting basic knowledge about a specific domain, and has been used in applications such as the representation of the meaning of natural language sentences in artificial intelligence applications. In a general setting the term has been used to refer to the use of a limited store of generally understood knowledge about a specific domain in the world, and has been applied to fields such as the knowledge based design of data schemas.[243]
In programming languages, name binding is the association of entities (data and/or code) withidentifiers.[244] An identifier bound to an object is said toreference that object.Machine languages have no built-in notion of identifiers, but name-object bindings as a service and notation for the programmer is implemented by programming languages. Binding is intimately connected withscoping, as scope determines which names bind to which objects – at which locations in the program code (lexically) and in which one of the possible execution paths (temporally). Use of an identifierid in a context that establishes a binding forid is called a binding (or defining) occurrence. In all other occurrences (e.g., in expressions, assignments, and subprogram calls), an identifier stands for what it is bound to; such occurrences are called applied occurrences.
A subtask ofinformation extraction that seeks to locate and classifynamed entity mentions inunstructured text into pre-defined categories such as the person names, organizations, locations,medical codes, time expressions, quantities, monetary values, percentages, etc.
A key concept ofSemantic Web architecture in which a set ofResource Description Framework statements (agraph) are identified using aURI,[245] allowing descriptions to be made of that set of statements such as context, provenance information or other suchmetadata. Named graphs are a simple extension of the RDF data model[246] through which graphs can be created but the model lacks an effective means of distinguishing between them once published on theWeb at large.
A software process that transforms structured data into plain-English content. It can be used to produce long-form content for organizations to automate custom reports, as well as produce custom content for a web or mobile application. It can also be used to generate short blurbs of text in interactive conversations (achatbot) which might even be read out loud by atext-to-speech system.
A subfield of computer science, information engineering, and artificial intelligence concerned with the interactions between computers and human (natural) languages, in particular how to program computers to process and analyze large amounts ofnatural language data.
All networks, including biological networks, social networks, technological networks (e.g., computer networks and electrical circuits) and more, can be represented asgraphs, which include a wide variety of subgraphs. One important local property of networks are so-called network motifs, which are defined as recurrent andstatistically significant sub-graphs or patterns.
An approach tomachine translation that uses a largeartificial neural network to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model.
Aneural network can refer to either aneural circuit of biologicalneurons (sometimes also called abiological neural network),or a network ofartificial neurons ornodes in the case of an artificial neural network.[248] Artificial neural networks are used for solvingartificial intelligence (AI) problems; they model connections of biological neurons as weights between nodes. A positive weight reflects an excitatory connection, while negative values mean inhibitory connections. All inputs are modified by a weight and summed. This activity is referred to as alinear combination. Finally, anactivation function controls theamplitude of the output. For example, an acceptable range of output is usually between 0 and 1, or it could be −1 and 1.
A direct communication pathway between an enhanced or wiredbrain and an external device. BCI differs fromneuromodulation in that it allows for bidirectional information flow. BCIs are often directed at researching, mapping, assisting, augmenting, or repairing human cognitive or sensory-motor functions.[251]
A basic unit of adata structure, such as alinked list ortree data structure. Nodes containdata and also may link to other nodes. Links between nodes are often implemented bypointers.
Nouvelle AI differs fromclassical AI by aiming to produce robots with intelligence levels similar to insects. Researchers believe that intelligence can emerge organically from simple behaviors as these intelligences interacted with the "real world", instead of using the constructed worlds which symbolic AIs typically needed to have programmed into them.[259]
Incomputational complexity theory, a problem is NP-complete when it can be solved by a restricted class ofbrute force search algorithms and it can be used to simulate any other problem with a similar algorithm. More precisely, each input to the problem should be associated with a set of solutions of polynomial length, whose validity can be tested quickly (inpolynomial time[261]), such that the output for any input is "yes" if the solution set is non-empty and "no" if it is empty.
Incomputational complexity theory, the defining property of a class of problems that are, informally, "at least as hard as the hardest problems in NP". A simple example of an NP-hard problem is thesubset sum problem.
The problem-solving principle that states that when presented with competinghypotheses that make the same predictions, one should select the solution with the fewest assumptions;[262] the principle is not meant to filter out hypotheses that make different predictions. The idea is attributed to the EnglishFranciscan friarWilliam of Ockham (c. 1287–1347), ascholastic philosopher andtheologian.
A method ofmachine learning in which data becomes available in a sequential order and is used to update the best predictor for future data at each step, as opposed to batch learning techniques which generate the best predictor by learning on the entire training data set at once. Online learning is a common technique used in areas of machine learning where it is computationally infeasible to train over the entire dataset, requiring the need ofout-of-core algorithms. It is also used in situations where it is necessary for the algorithm to dynamically adapt to new patterns in the data, or when the data itself is generated as a function of time.
The automatic or semi-automatic creation ofontologies, including extracting the correspondingdomain's terms and the relationships between theconcepts that these terms represent from acorpus of natural language text, and encoding them with anontology language for easy retrieval.
The for-profit corporation OpenAI LP, whoseparent organization is the non-profit organization OpenAI Inc[263] that conducts research in the field ofartificial intelligence (AI) with the stated aim to promote and developfriendly AI in such a way as to benefit humanity as a whole.
"The production of an analysis that corresponds too closely or exactly to a particular set of data, and may therefore fail to fit to additional data or predict future observations reliably".[267] In other words, an overfitted model memorizes training data details but cannotgeneralize to new data. Conversely, anunderfitted model is too simple to capture the complexity of the training data.
A generalization of aMarkov decision process (MDP). A POMDP models an agent decision process in which it is assumed that the system dynamics are determined by an MDP, but the agent cannot directly observe the underlying state. Instead, it must maintain a probability distribution over the set of possible states, based on a set of observations and observation probabilities, and the underlying MDP.
A computational method thatoptimizes a problem byiteratively trying to improve acandidate solution with regard to a given measure of quality. It solves a problem by having a population of candidate solutions, here dubbedparticles, and moving these particles around in thesearch-space according to simplemathematical formulae over the particle's position and velocity. Each particle's movement is influenced by its local best known position, but is also guided toward the best known positions in the search-space, which are updated as better positions are found by other particles. This is expected to move the swarm toward the best solutions.
The plotting, by a computer application, of the shortest route between two points. It is a more practical variant onsolving mazes. This field of research is based heavily onDijkstra's algorithm for finding a shortest path on aweighted graph.
Concerned with the automatic discovery of regularities in data through the use of computer algorithms and with the use of these regularities to take actions such as classifying the data into different categories.[268]
A collection offormal systems used inmathematics,philosophy,linguistics, andcomputer science. First-order logic usesquantified variables over non-logical objects and allows the use of sentences that contain variables, so that rather than propositions such asSocrates is a man one can have expressions in the form "there exists x such that x is Socrates and x is a man" andthere exists is a quantifier whilex is a variable.[182] This distinguishes it frompropositional logic, which does not use quantifiers orrelations;[269] in this sense, propositional logic is the foundation of first-order logic.
A statistical procedure that uses anorthogonal transformation to convert a set of observations of possibly correlated variables (entities each of which takes on various numerical values) into a set of values oflinearly uncorrelated variables called principal components. This transformation is defined in such a way that the first principal component has the largest possiblevariance (that is, accounts for as much of the variability in the data as possible), and each succeeding component, in turn, has the highest variance possible under the constraint that it isorthogonal to the preceding components. The resulting vectors (each being alinear combination of the variables and containingn observations) are an uncorrelatedorthogonal basis set. PCA is sensitive to the relative scaling of the original variables.
A principle coined byKarl R. Popper in his Harvard Lecture of 1963, and published in his bookMyth of Framework.[272] It is related to what he called the 'logic of the situation' in anEconomica article of 1944/1945, published later in his bookThe Poverty of Historicism.[273] According to Popper's rationality principle, agents act in the most adequate way according to the objective situation. It is an idealized conception of human behavior which he used to drive his model ofsituational logic.
Aprogramming paradigm in whichprobabilistic models are specified and inference for these models is performed automatically.[274] It represents an attempt to unify probabilistic modeling and traditional general-purpose programming in order to make the former easier and more widely applicable.[275][276] It can be used to create systems that help make decisions in the face of uncertainty. Programming languages used for probabilistic programming are referred to as "Probabilistic programming languages" (PPLs).
A computer program typically used to provide some form of AI, which consists primarily of a set of rules about behavior, but also includes the mechanism necessary to follow those rules as the system responds to states of the world.
A branch oflogic which deals withpropositions (which can be true or false) and argument flow. Compound propositions are formed by connecting propositions bylogical connectives. The propositions without logical connectives are called atomic propositions. Unlikefirst-order logic, propositional logic does not deal with non-logical objects, predicates about them, or quantifiers. However, all the machinery of propositional logic is included in first-order logic and higher-order logics. In this sense, propositional logic is the foundation of first-order logic and higher-order logic.
In philosophy and artificial intelligence (especiallyknowledge-based systems), the qualification problem is concerned with the impossibility of listingall of thepreconditions required for a real-world action to have its intended effect.[290][291] It might be posed ashow to deal with the things that prevent me from achieving my intended result. It is strongly connected to, and opposite theramification side of, theframe problem.[290]
Inlogic, quantification specifies the quantity of specimens in thedomain of discourse that satisfy anopen formula. The two most common quantifiers mean "for all" and "there exists". For example, in arithmetic, quantifiers allow one to say that the natural numbers go on forever, by writing thatfor all n (where n is a natural number), there is another number (say, the successor of n) which is one bigger than n.
Query languages ordata query languages (DQLs) arecomputer languages used to make queries indatabases andinformation systems. Broadly, query languages can be classified according to whether they are database query languages orinformation retrieval query languages. The difference is that a database query language attempts to give factual answers to factual questions, while an information retrieval query language attempts to find documents containing information that is relevant to an area of inquiry.
A set of statistical processes for estimating the relationships between a dependent variable (often called the outcome or response variable, orlabel inmachine learning) and one or more error-free independent variables (often called regressors, predictors, covariates, explanatory variables, orfeatures). The most common form of regression analysis is linear regression, in which one finds the line (or a more complex linear combination) that most closely fits the data according to a specific mathematical criterion.
An area ofmachine learning concerned with howsoftware agents ought to takeactions in an environment so as to maximize some notion of cumulative reward. Reinforcement learning is one of three basic machine learning paradigms, alongsidesupervised andunsupervised learning. It differs from supervised learning in that labelled input/output pairs need not be presented, and sub-optimal actions need not be explicitly corrected. Instead the focus is finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge).[305]
A technique that involve training a "reward model" to predict how humans rate the quality of generated content, and then training agenerative AI model to satisfy this reward model viareinforcement learning. It can be used for example to make the generative AI model more truthful or less harmful.[306]
A framework for computation that may be viewed as an extension ofneural networks.[307] Typically an input signal is fed into a fixed (random)dynamical system called areservoir and the dynamics of the reservoir map the input to a higher dimension. Then a simplereadout mechanism is trained to read the state of the reservoir and map it to the desired output. The main benefit is that training is performed only at the readout stage and the reservoir is fixed.Liquid-state machines[308] andecho state networks[309] are two major types of reservoir computing.[310]
Apattern matchingalgorithm for implementingrule-based systems. The algorithm was developed to efficiently apply manyrules or patterns to many objects, orfacts, in aknowledge base. It is used to determine which of the system's rules should fire based on its data store, its facts.
Incomputer science, a rule-based system is used to store and manipulate knowledge to interpret information in a useful way. It is often used in artificial intelligence applications and research. Normally, the termrule-based system is applied to systems involving human-crafted or curated rule sets. Rule-based systems constructed using automatic rule inference, such asrule-based machine learning, are normally excluded from this system type.
Inmathematical logic, satisfiability andvalidity are elementary concepts ofsemantics. Aformula issatisfiable if it is possible to find aninterpretation (model) that makes the formula true.[312] A formula isvalid if all interpretations make the formula true. The opposites of these concepts are unsatisfiability and invalidity, that is, a formula isunsatisfiable if none of the interpretations make the formula true, andinvalid if some such interpretation makes the formula false. These four concepts are related to each other in a manner exactly analogous toAristotle'ssquare of opposition.
Allows for queries and analytics of associative andcontextual nature. Semantic queries enable the retrieval of both explicitly and implicitly derived information based on syntactic, semantic and structural information contained in data. They are designed to deliver precise results (possibly the distinctive selection of one single piece of information) or to answer more fuzzy and wide-open questions throughpattern matching anddigital reasoning.
Inprogramming language theory, semantics is the field concerned with the rigorous mathematical study of the meaning ofprogramming languages. It does so by evaluating the meaning ofsyntactically validstrings defined by a specific programming language, showing the computation involved. In such a case that the evaluation would be of syntactically invalid strings, the result would be non-computation. Semantics describes the processes a computer follows when executing a program in that specific language. This can be shown by describing the relationship between the input and output of a program, or an explanation of how the program will be executed on a certainplatform, hence creating amodel of computation.
Amachine learning training paradigm characterized by using a combination of a small amount of human-labeled data (used exclusively insupervised learning), followed by a large amount of unlabeled data (used exclusively inunsupervised learning).
The combining ofsensory data or data derived from disparate sources such that the resultinginformation has less uncertainty than would be possible when these sources were used individually.
An area ofsupervised learning closely related toclassification andregression, but the goal is to learn from a similarity function that measures how similar or related two objects are. It has applications inranking, inrecommendation systems, visual identity tracking, face verification, and speaker verification.
In artificial intelligence research, the situated approach builds agents that are designed to behave effectively successfully in their environment. This requires designing AI "from the bottom-up" by focussing on the basic perceptual and motor skills required to survive. The situated approach gives a much lower priority to abstract reasoning or problem-solving skills.
An area of artificial intelligence which draws from the fields ofcomputer science,cognitive science, andcognitive psychology. The theoretic goal—on the cognitive side—involves representing and reasoning spatial-temporal knowledge in mind. The applied goal—on the computing side—involves developing high-level control systems of automata fornavigating and understanding time and space.
Afeature learning method aimed at finding asparse representation of the input data in the form of a linear combination of basic elements as well as those basic elements themselves.
An interdisciplinary subfield ofcomputational linguistics that develops methodologies and technologies that enables the recognition andtranslation of spoken language into text by computers. It is also known as automatic speech recognition (ASR), computer speech recognition or speech to text (STT). It incorporates knowledge and research in thelinguistics,computer science, andelectrical engineering fields.
Ininformation technology andcomputer science, a program is described as stateful if it is designed to remember preceding events or user interactions;[321] the remembered information is called the state of the system.
Inmachine learning andstatistics, classification is the problem of identifying to which of a set ofcategories (sub-populations) a new observation belongs, on the basis of atraining set of data containing observations (or instances) whose category membership is known. Examples are assigning a given email to the"spam" or "non-spam" class, and assigning a diagnosis to a given patient based on observed characteristics of the patient (sex, blood pressure, presence or absence of certain symptoms, etc.). Classification is an example ofpattern recognition.
Anyoptimizationmethod that generates and usesrandom variables. For stochastic problems, the random variables appear in the formulation of the optimization problem itself, which involves randomobjective functions or random constraints. Stochastic optimization methods also include methods with random iterates. Some stochastic optimization methods use random iterates to solve stochastic problems, combining both meanings of stochastic optimization.[324] Stochastic optimization methods generalizedeterministic methods for deterministic problems.
An approach used incomputer science as asemantic component ofnatural language understanding. Stochastic models generally use the definition of segments of words as basic semantic units for the semantic models, and in some cases involve a two layered approach.[325]
A person who has accumulated great knowledge in a particular field or topic, demonstrated by the person's degree, licensure, and/or through years of professional experience with the subject.
A hypotheticalagent that possessesintelligence far surpassing that of thebrightest and mostgifted human minds. Superintelligence may also refer to a property of problem-solving systems (e.g., superintelligent language translators or engineering assistants) whether or not these high-level intellectual competencies are embodied in agents that act within the physical world. A superintelligence may or may not be created by anintelligence explosion and be associated with atechnological singularity.
Themachine learning task of learning a function that maps an input to an output based on example input-output pairs.[326] It infers a function fromlabeledtraining data consisting of a set oftraining examples.[327] In supervised learning, each example is apair consisting of an input object (typically a vector) and a desired output value (also called thesupervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which can be used for mapping new examples. An optimal scenario will allow for the algorithm to correctly determine the class labels for unseen instances. This requires the learning algorithm togeneralize from the training data to unseen situations in a "reasonable" way (seeinductive bias).
The term for the collection of all methods inartificial intelligence research that are based on high-level "symbolic" (human-readable) representations of problems,logic, andsearch.
An alternative term forartificial intelligence which emphasizes that the intelligence of machines need not be an imitation or in any way artificial; it can be a genuine form of intelligence.[330][331]
Ahypothetical point in the future when technological growth becomes uncontrollable and irreversible, resulting in unfathomable changes to human civilization.[332][333][334]
Aheuristic for choosing actions that addresses the exploration-exploitation dilemma in themulti-armed bandit problem. It consists in choosing the action that maximizes the expected reward with respect to a randomly drawn belief.[340][341]
Thecomputational complexity that describes the amount of time it takes to run analgorithm. Time complexity is commonly estimated by counting the number of elementary operations performed by the algorithm, supposing that each elementary operation takes a fixed amount of time to perform. Thus, the amount of time taken and the number of elementary operations performed by the algorithm are taken to differ by at most aconstant factor.
Amachine learning technique in which knowledge learned from a task is reused in order to boost performance on a related task.[342] For example, forimage classification, knowledge gained while learning torecognize cars could be applied when trying to recognize trucks.
Intheoretical computer science, a transition system is a concept used in the study ofcomputation. It is used to describe the potential behavior ofdiscrete systems. It consists ofstates and transitions between states, which may be labeled with labels chosen from a set; the same label may appear on more than one transition. If the label set is asingleton, the system is essentially unlabeled, and a simpler definition that omits the labels is possible.
A form ofgraph traversal and refers to the process of visiting (checking and/or updating) each node in atree data structure, exactly once. Such traversals are classified by the order in which the nodes are visited.
Incomputational complexity theory, the language TQBF is aformal language consisting of the true quantified Boolean formulas. A (fully) quantified Boolean formula is a formula inquantifiedpropositional logic where every variable is quantified (orbound), using eitherexistential oruniversal quantifiers, at the beginning of the sentence. Such a formula is equivalent to either true or false (since there are nofree variables). If such a formula evaluates to true, then that formula is in the language TQBF. It is also known as QSAT (QuantifiedSAT).
A test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human, developed byAlan Turing in 1950. Turing proposed that a human evaluator wouldjudge natural language conversations between a human and a machine designed to generate human-like responses. The evaluator would be aware that one of the two partners in conversation is a machine, and all participants would be separated from one another. The conversation would be limited to a text-only channel such as acomputer keyboard andscreen so the result would not depend on the machine's ability to render words as speech.[349] If the evaluator cannot reliably tell the machine from the human, the machine is said to have passed the test. The test results do not depend on the machine's ability to give correct answers to questions, only how closely its answers resemble those a human would give.
Inprogramming languages, a set of rules that assigns a property calledtype to the various constructs of acomputer program, such asvariables,expressions,functions, ormodules.[350] These types formalize and enforce the otherwise implicit categories the programmer uses foralgebraic data types, data structures, or other components (e.g. "string", "array of float", "function returning boolean"). The main purpose of a type system is to reduce possibilities forbugs in computer programs[351] by defininginterfaces between different parts of a computer program, and then checking that the parts have been connected in a consistent way. This checking can happen statically (atcompile time), dynamically (atrun time), or as a combination of static and dynamic checking. Type systems have other purposes as well, such as expressing business rules, enabling certain compiler optimizations, allowing formultiple dispatch, providing a form of documentation, etc.
A representation of a word innatural language processing. Typically, the representation is areal-valued vector that encodes the meaning of the word in such a way that words that are closer in the vector space are expected to be similar in meaning.[362]
^"Retroduction".Commens – Digital Companion to C. S. Peirce. Mats Bergman, Sami Paavola & João Queiroz. Archived fromthe original on 5 July 2022. Retrieved24 August 2014.
^Abraham, A. (2005), "Adaptation of Fuzzy Inference System Using Neural Learning", in Nedjah, Nadia; De Macedo Mourelle, Luiza (eds.),Fuzzy Systems Engineering: Theory and Practice, Studies in Fuzziness and Soft Computing, vol. 181, Germany: Springer Verlag, pp. 53–83,CiteSeerX10.1.1.161.6135,doi:10.1007/11339366_3,ISBN978-3-540-25322-8
^Tao, Jianhua; Tieniu Tan (2005). "Affective Computing: A Review".Affective Computing and Intelligent Interaction. Vol. LNCS 3784. Springer. pp. 981–995.doi:10.1007/11573548.
^El Kaliouby, Rana (November–December 2017)."We Need Computers with Empathy".Technology Review. Vol. 120, no. 6. p. 8. Archived fromthe original on 7 July 2018. Retrieved6 November 2018.
^Shapiro, Stuart C. (1992).Artificial Intelligence In Stuart C. Shapiro (Ed.),Encyclopedia of Artificial Intelligence (Second Edition, pp. 54–57). New York: John Wiley. (Section 4 is on "AI-Complete Tasks".)
Poole, Mackworth & Goebel 1998,p. 1, which provides the version that is used in this article. Note that they use the term "computational intelligence" as a synonym for artificial intelligence.
^Gehrig, Stefan K.; Stein, Fridtjof J. (1999).Dead reckoning and cartography using stereo vision for an automated car. IEEE/RSJ International Conference on Intelligent Robots and Systems. Vol. 3. Kyongju. pp. 1507–1512.doi:10.1109/IROS.1999.811692.ISBN0-7803-5184-3.
^Ioffe, Sergey; Szegedy, Christian (2015). "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift".arXiv:1502.03167 [cs.LG].
^Pham DT, Ghanbarzadeh A, Koc E, Otri S, Rahim S and Zaidi M. The Bees Algorithm. Technical Note, Manufacturing Engineering Centre, Cardiff University, UK, 2005.
^Pham, D. T.; Castellani, M. (2014). "Benchmarking and comparison of nature-inspired population-based continuous optimisation algorithms".Soft Computing.18 (5):871–903.doi:10.1007/s00500-013-1104-9.S2CID35138140.
^Erman, L. D.; Hayes-Roth, F.; Lesser, V. R.; Reddy, D. R. (1980). "The Hearsay-II Speech-Understanding System: Integrating Knowledge to Resolve Uncertainty".ACM Computing Surveys.12 (2): 213.doi:10.1145/356810.356816.S2CID118556.
^Corkill, Daniel D. (September 1991)."Blackboard Systems"(PDF).AI Expert.6 (9):40–47. Archived fromthe original(PDF) on 16 April 2012. Retrieved5 July 2022.
^*Nii, H. Yenny (1986).Blackboard Systems(PDF) (Technical report). Department of Computer Science, Stanford University. STAN-CS-86-1123. Retrieved12 April 2013.
^Fisher, Douglas H. (July 1987). "Improving inference through conceptual clustering".Proceedings of the 1987 AAAI Conferences. AAAI Conference. Seattle Washington. pp. 461–465.
^Iba, William; Langley, Pat (27 January 2011). "Cobweb models of categorization and probabilistic concept formation". In Pothos, Emmanuel M.; Wills, Andy J. (eds.).Formal approaches in categorization. Cambridge: Cambridge University Press. pp. 253–273.ISBN978-0-521-19048-0.
^Cognitive science is an interdisciplinary field of researchers from Linguistics, psychology, neuroscience, philosophy, computer science, and anthropology that seek to understand the mind.How We Learn: Ask the Cognitive Scientist
^Schrijver, Alexander (February 1, 2006). A Course in Combinatorial Optimization (PDF), page 1.
^HAYKIN, S. Neural Networks – A Comprehensive Foundation. Second edition. Pearson Prentice Hall: 1999.
^Hulstijn, J, and Nijholt, A. (eds.). Proceedings of the International Workshop on Computational Humor. Number 12 in Twente Workshops on Language Technology, Enschede, Netherlands. University of Twente, 1996.
^Trappenberg, Thomas P. (2002). Fundamentals of Computational Neuroscience. United States: Oxford University Press Inc. p. 1.ISBN978-0-19-851582-1.
^What is computational neuroscience? Patricia S. Churchland, Christof Koch, Terrence J. Sejnowski. in Computational Neuroscience pp.46–55. Edited by Eric L. Schwartz. 1993. MIT Press"Computational Neuroscience Edited by Eric L. Schwartz". 26 August 1993. Archived fromthe original on 4 June 2011. Retrieved11 June 2009.
^Kramer, GJE; Grierson, DE (1989). "Computer automated design of structures under dynamic loads".Computers & Structures.32 (2):313–325.doi:10.1016/0045-7949(89)90043-6.
^Moharrami, H; Grierson, DE (1993). "Computer-Automated Design of Reinforced Concrete Frameworks".Journal of Structural Engineering.119 (7):2036–2058.doi:10.1061/(asce)0733-9445(1993)119:7(2036).
^Xu, L; Grierson, DE (1993). "Computer-Automated Design of Semirigid Steel Frameworks".Journal of Structural Engineering.119 (6):1740–1760.doi:10.1061/(asce)0733-9445(1993)119:6(1740).
^Barsan, GM; Dinsoreanu, M, (1997). Computer-automated design based on structural performance criteria, Mouchel Centenary Conference on Innovation in Civil and Structural Engineering, Aug 19-21, Cambridge England, Innovation in Civil and Structural Engineering, 167–172
^Li, Yun (1996). "Genetic algorithm automated approach to the design of sliding mode control systems".International Journal of Control.63 (4):721–739.doi:10.1080/00207179608921865.
^Barsan, GM, (1995) Computer-automated design of semirigid steel frameworks according to EUROCODE-3, Nordic Steel Construction Conference 95, JUN 19–21, 787–794
^Dana H. Ballard; Christopher M. Brown (1982). Computer Vision. Prentice Hall.ISBN0-13-165316-4.
^Huang, T. (1996-11-19). Vandoni, Carlo, E, ed. Computer Vision : Evolution And Promise (PDF). 19th CERN School of Computing. Geneva: CERN. pp. 21–25.doi:10.5170/CERN-1996-008.21.ISBN978-9290830955.
^Milan Sonka; Vaclav Hlavac; Roger Boyle (2008). Image Processing, Analysis, and Machine Vision. Thomson.ISBN0-495-08252-X.
^Garson, James (27 November 2018). Zalta, Edward N. (ed.).The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University – via Stanford Encyclopedia of Philosophy.
^Zhang, Wei (1988). "Shift-invariant pattern recognition neural network and its optical architecture". Proceedings of annual conference of the Japan Society of Applied Physics.
^Solomonoff, R.J.The Time Scale of Artificial Intelligence; Reflections on Social Effects, Human Systems Management, Vol 5 1985, Pp 149–153
^Moor, J., The Dartmouth College Artificial Intelligence Conference: The Next Fifty years, AI Magazine, Vol 27, No., 4, Pp. 87–9, 2006
^Kline, Ronald R., Cybernetics, Automata Studies and the Dartmouth Conference on Artificial Intelligence, IEEE Annals of the History of Computing, October–December, 2011, IEEE Computer Society
^Huang; Green; Loo, "Datalog and Emerging applications",SIGMOD 2011(PDF), UC Davis, archived fromthe original(PDF) on 1 July 2022, retrieved5 July 2022.
^Steele, Katie and Stefánsson, H. Orri, "Decision Theory", The Stanford Encyclopedia of Philosophy (Winter 2015 Edition), Edward N. Zalta (ed.), URL =[1]
^Lloyd, J.W.,Practical Advantages of Declarative Programming
^abOstrow, Adam (5 March 2011)."Roger Ebert's Inspiring Digital Transformation". Mashable Entertainment. Retrieved12 September 2011.With the help of his wife, two colleagues and the Alex-equipped MacBook that he uses to generate his computerized voice, famed film critic Roger Ebert delivered the final talk at the TED conference on Friday in Long Beach, California....
^Lee, Jennifer (7 March 2011)."Roger Ebert Tests His Vocal Cords, and Comedic Delivery".The New York Times. Retrieved12 September 2011.Now perhaps, there is the Ebert Test, a way to see if a synthesized voice can deliver humor with the timing to make an audience laugh.... He proposed the Ebert Test as a way to gauge the humanness of a synthesized voice.
^"Roger Ebert's Inspiring Digital Transformation". Tech News. 5 March 2011. Archived fromthe original on 25 March 2011. Retrieved12 September 2011.Meanwhile, the technology that enables Ebert to "speak" continues to see improvements – for example, adding more realistic inflection for question marks and exclamation points. In a test of that, which Ebert called the "Ebert test" for computerized voices,
^Vikhar, P. A. (2016). "Evolutionary algorithms: A critical review and its future prospects".2016 International Conference on Global Trends in Signal Processing, Information Computing and Communication (ICGTSPICC). Jalgaon, 2016, pp. 261–265. pp. 261–265.doi:10.1109/ICGTSPICC.2016.7955308.ISBN978-1-5090-0467-6.S2CID22100336.
^Sardar, Z (2010). "The Namesake: Futures; futures studies; futurology; futuristic; Foresight – What's in a name?".Futures.42 (3):177–184.doi:10.1016/j.futures.2009.11.001.
^Pedrycz, Witold (1993).Fuzzy control and fuzzy systems (2 ed.). Research Studies Press Ltd.
^Hájek, Petr (1998).Metamathematics of fuzzy logic (4 ed.). Springer Science & Business Media.
^D. Dubois and H. Prade (1988) Fuzzy Sets and Systems. Academic Press, New York.
^Myerson, Roger B. (1991).Game Theory: Analysis of Conflict, Harvard University Press, p. 1. Chapter-preview links, pp.vii–xi.
^Pell, Barney (1992). H. van den Herik; L. Allis (eds.)."Metagame: a new challenge for games and learning" [Heuristic programming in artificial intelligence 3–the third computerolympiad](PDF). Ellis-Horwood. Archived fromthe original(PDF) on 17 February 2020. Retrieved13 June 2020.
^Genesereth, Michael; Love, Nathaniel; Pell, Barney (15 June 2005). "General Game Playing: Overview of the AAAI Competition".AI Magazine.26 (2): 62.doi:10.1609/aimag.v26i2.1813.ISSN2371-9621.
^Gluck, Mark A.; Mercado, Eduardo; Myers, Catherine E. (2011).Learning and memory: from brain to behavior (2nd ed.). New York: Worth Publishers. p. 209.ISBN978-1-4292-4014-7.
^Mohri, M., Rostamizadeh A., Talwakar A., (2018)Foundations of Machine learning, 2nd ed., Boston: MIT Press
^Y S. Abu-Mostafa, M.Magdon-Ismail, and H.-T. Lin (2012) Learning from Data, AMLBook Press.ISBN978-1600490064
^Andrej Karpathy; Pieter Abbeel; Greg Brockman; Peter Chen; Vicki Cheung; Yan Duan; Ian Goodfellow; Durk Kingma; Jonathan Ho; Rein Houthooft; Tim Salimans; John Schulman; Ilya Sutskever; Wojciech Zaremba (16 June 2016)."Generative models".OpenAI.
^Trudeau, Richard J. (1993).Introduction to Graph Theory (Corrected, enlarged republication. ed.). New York: Dover Pub. p. 19.ISBN978-0-486-67870-2. Retrieved8 August 2012.A graph is an object consisting of two sets called itsvertex set and itsedge set.
^E. K. Burke, E. Hart,G. Kendall, J. Newall, P. Ross, and S. Schulenburg, Hyper-heuristics: An emerging direction in modern search technology, Handbook of Metaheuristics (F. Glover and G. Kochenberger, eds.), Kluwer, 2003, pp. 457–474.
^P. Ross, Hyper-heuristics, Search Methodologies: Introductory Tutorials in Optimization and Decision Support Techniques (E. K. Burke andG. Kendall, eds.), Springer, 2005, pp. 529–556.
^Ozcan, E.; Bilgin, B.; Korkmaz, E. E. (2008). "A Comprehensive Analysis of Hyper-heuristics".Intelligent Data Analysis.12 (1):3–23.doi:10.3233/ida-2008-12102.
^Oudeyer, Pierre-Yves; Kaplan, Frederic (2008). "How can we define intrinsic motivation?".Proc. of the 8th Conf. on Epigenetic Robotics. Vol. 5. pp. 29–31.
^Schank, Roger; Robert Abelson (1977).Scripts, Plans, Goals, and Understanding: An Inquiry Into Human Knowledge Structures. Lawrence Erlbaum Associates, Inc.
^Siegelmann, Hava T.; Sontag, Eduardo D. (1992). "On the computational power of neural nets".Proceedings of the fifth annual workshop on Computational learning theory. Vol. COLT '92. pp. 440–449.doi:10.1145/130385.130432.ISBN978-0-89791-497-0.S2CID207165680.{{cite book}}:|work= ignored (help)
^Herbert B. Enderton, 2001, A Mathematical Introduction to Logic Second Edition Enderton:110, Harcourt Academic Press, Burlington MA,ISBN978-0-12-238452-3.
^Bishop, Christopher M. (2006). Pattern Recognition and Machine Learning (PDF). Springer. p. vii. Pattern recognition has its origins in engineering, whereas machine learning grew out of computer science. However, these activities can be viewed as two facets of the same field, and together they have undergone substantial development over the past ten years.
^Nyce, Charles (2007),Predictive Analytics White Paper(PDF), American Institute for Chartered Property Casualty Underwriters/Insurance Institute of America, p. 1
^Clocksin, William F.; Mellish, Christopher S. (2003).Programming in Prolog. Berlin; New York: Springer-Verlag.ISBN978-3-540-00678-7.
^Bratko, Ivan (2012).Prolog programming for artificial intelligence (4th ed.). Harlow, England; New York: Addison Wesley.ISBN978-0-321-41746-6.
^Covington, Michael A. (1994).Natural language processing for Prolog programmers. Englewood Cliffs, N.J.: Prentice Hall.ISBN978-0-13-629213-5.
^Lloyd, J. W. (1984). Foundations of logic programming. Berlin: Springer-Verlag.ISBN978-3-540-13299-8.
^Kuhlman, Dave. "A Python Book: Beginning Python, Advanced Python, and Python Exercises". Section 1.1. Archived from the original (PDF) on 23 June 2012.
^Mannes, John."Facebook and Microsoft collaborate to simplify conversions from PyTorch to Caffe2".TechCrunch. Retrieved18 December 2017.FAIR is accustomed to working with PyTorch – a deep learning framework optimized for achieving state of the art results in research, regardless of resource constraints. Unfortunately in the real world, most of us are limited by the computational capabilities of our smartphones and computers.
^Thielscher, Michael (September 2001). "The Qualification Problem: A solution to the problem of anomalous models".Artificial Intelligence.131 (1–2):1–37.doi:10.1016/S0004-3702(01)00131-X.
^ R language and environmentHornik, Kurt (4 October 2017)."R FAQ".The Comprehensive R Archive Network. 2.1 What is R?. Retrieved6 August 2018.R FoundationHornik, Kurt (4 October 2017)."R FAQ".The Comprehensive R Archive Network. 2.13 What is the R Foundation?. Retrieved6 August 2018.The R Core Team asks authors who use R in their data analysis to cite the software using:R Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URLhttps://R-project.org/.
^Vance, Ashlee (6 January 2009)."Data Analysts Captivated by R's Power".The New York Times. Retrieved6 August 2018.R is also the name of a popular programming language used by a growing number of data analysts inside corporations and academia. It is becoming their lingua franca...
^Ho, Tin Kam (1995). Random Decision Forests (PDF). Proceedings of the 3rd International Conference on Document Analysis and Recognition, Montreal, QC, 14–16 August 1995. pp. 278–282. Archived from the original (PDF) on 17 April 2016. Retrieved 5 June 2016.
^Li, Xiangang; Wu, Xihong (15 October 2014). "Constructing Long Short-Term Memory based Deep Recurrent Neural Networks for Large Vocabulary Speech Recognition".arXiv:1410.4281 [cs.CL].
^Schrauwen, Benjamin,David Verstraeten, andJan Van Campenhout. "An overview of reservoir computing: theory, applications, and implementations." Proceedings of the European Symposium on Artificial Neural Networks ESANN 2007, pp. 471–482.
^Jaeger, Herbert, "The echo state approach to analyzing and training recurrent neural networks." Technical Report 154 (2001), German National Research Center for Information Technology.
^Beni, G.; Wang, J. (1993). "Swarm Intelligence in Cellular Robotic Systems".Proceed. NATO Advanced Workshop on Robots and Biological Systems, Tuscany, Italy, June 26–30 (1989). pp. 703–712.doi:10.1007/978-3-642-58069-7_38.ISBN978-3-642-63461-1.
^Pellionisz, A.; Llinás, R. (1985). "Tensor Network Theory Of The Metaorganization Of Functional Geometries In The Central Nervous System".Neuroscience.16 (2):245–273.doi:10.1016/0306-4522(85)90001-6.PMID4080158.S2CID10747593.
^"TensorFlow: Open source machine learning" "It is machine learning software being used for various kinds of perceptual and language understanding tasks" — Jeffrey Dean, minute 0:47 / 2:17 from YouTube clip
^Sipser, Michael (2013).Introduction to the Theory of Computation 3rd. Cengage Learning.ISBN978-1-133-18779-0.central areas of the theory of computation: automata, computability, and complexity. (Page 1)
^Thompson, William R (1933). "On the likelihood that one unknown probability exceeds another in view of the evidence of two samples".Biometrika.25 (3–4):285–294.doi:10.1093/biomet/25.3-4.285.
^Russo, Daniel J.; Van Roy, Benjamin; Kazerouni, Abbas; Osband, Ian; Wen, Zheng (2018). "A Tutorial on Thompson Sampling".Foundations and Trends in Machine Learning.11 (1):1–96.arXiv:1707.02038.doi:10.1561/2200000070.S2CID3929917.
^Minsky 1967:107 "In his 1936 paper, A. M. Turing defined the class of abstract machines that now bear his name. A Turing machine is a finite-state machine associated with a special kind of environment – its tape – in which it can store (and later recover) sequences of symbols," also Stone 1972:8 where the word "machine" is in quotation marks.
^Stone 1972:8 states "This "machine" is an abstract mathematical model", also cf. Sipser 2006:137ff that describes the "Turing machine model". Rogers 1987 (1967):13 refers to "Turing's characterization", Boolos Burgess and Jeffrey 2002:25 refers to a "specific kind of idealized machine".
^Sipser 2006:137 "A Turing machine can do everything that a real computer can do".
^Turing originally suggested ateleprinter, one of the few text-only communication systems available in 1950. (Turing 1950, p. 433)
^Pierce 2002, p. 1: "A type system is a tractable syntactic method for proving the absence of certain program behaviors by classifying phrases according to the kinds of values they compute."
^Cardelli 2004, p. 1: "The fundamental purpose of a type system is to prevent the occurrence of execution errors during the running of a program."
^Hinton, Jeffrey; Sejnowski, Terrence (1999).Unsupervised Learning: Foundations of Neural Computation. MIT Press.ISBN978-0-262-58168-4.
Abran, Alain; Moore, James W.; Bourque, Pierre; Dupuis, Robert; Tripp, Leonard L. (2004).Guide to the Software Engineering Body of Knowledge. IEEE.ISBN978-0-7695-2330-9.
^polynomial time refers to how quickly the number of operations needed by an algorithm, relative to the size of the problem, grows. It is therefore a measure of efficiency of an algorithm.
_artificial_intelligence_icon.png){kind=icon}
