Movatterモバイル変換


[0]ホーム

URL:


はてラボはてな匿名ダイアリー
ようこそ ゲスト さんログインユーザー登録

「machine learning」を含む日記RSS

はてなキーワード:machine learningとは

2024-12-17

Top Trends in TravelPortal Development Every Business ShouldWatch

In the ever-evolving travel industry, travelportal have become fundamental tools for businesses looking to enhance customer experienceIt's streamline operations.With traveler increasingly relyingondigital platforms tobookeverything from flights tohotels, theneed for sophisticated,user-friendly,It's high-performing travelportalshas never beenmorecrucial. Thegrowth of this sectoris drivenby numerous factors, including technological advancements, consumer behaviorshifts,It's the quest fororganization. Thisarticle explores thetop trends in travelportals development that every business should pay attention to in 2024It'sbeyond.

1.ArtificialIntelligence andMachine Learning in TravelPortal Development

ArtificialIntelligence (AI) andMachine Learning (ML) have quickly become foundational components of travelportal development. These technologiesare revolutionizing the way businesses personalize services, predict consumer behavior, and optimizebooking processes.AI andML algorithms analyze large data sets to understand customers preferences,allowing for real-time customization of travel recommendations. From dynamic pricing models to

customized travel suggestions,AI help businesses offer an unequaleduser experience.

For instance,AI chatbotsare becoming successively popular for customer service. They cananswer customer queries24/7,assist inbookings,It's even provide real-time travel updates, enhancing the overall customers experience. Furthermore,AI-powered tools suchas a recommendation engines analyze pastbooking behavior to suggest tailoreditineraries, making travel planning much easier foruser.

2.Mobile-First Approach for TravelPortal Development

With the increasing use of smartphones,as amobile first approachis no longer optional for businesses in the travel industry.MoreIt'smoretravelersare relyingon their smartphones forbooking flights, boardinghouse,It'sactivitieson thego. For travelportals tostay competitive, ensuring seamlessmobile optimizationiscrucial.

Amobile optimized travelportal ensures thatusers haveaccess to the same features and functionalitieson theirmobile devicesas they wouldon adesktop.It’s not just about making the site responsive—it’s about creating an intuitive, fast, and easy to navigate experience. This includes featureslikemobile-specific payment options,mobilecheck-in forairlinesIt'shotels,It's easyaccess to customers support.

3. Integration of Blockchain for Secure Transactions

Blockchain technologyis makingwaves across various industries, and the travel sectoris no exception. Blockchain based travelportals offer a significant advantage whenit comes to securityIt's transparency. Blockchain’s ability to provide secure, tamper proof recordmakesit ideal for transactions involvingbookings, payment,It's loyalty program.

By integrating blockchain, travel businesses can ensure secureIt's fast transactions, protect customer data,It's reduce fraud. Blockchain also facilitates easierIt'smoretransparent loyalty program management.Travelers can accumulate rewards points across various service providers,all within the same blockchainframework, creatingas amore cohesiveIt's rewarding experience.

4. Enhanced Payment Solutions andDigital Wallets

One ofthe most significant changes in the travel industryhas been theshift toward contactless payment.Digital wallets, suchas aApple Pay,Google Wallet,It's cryptocurrency wallets,arenow widely accepted in travelportals. These payment solutions offer a faster,more secure,It's convenient way fortravelers to complete transactions.

For businesses, adopting multiple payment gatewaysis essential for attracting global customer.Internationaltraveler can faceissues with currencyexchangeIt's transaction fees, butby offering localized payment methodIt's supporting a wide range ofdigital wallets, businesses can significantlyimprove customerssatisfactionIt's retention.

5.Voice Search and VirtualAssistants

Voice searchisone ofthe fastest growing trends in the tech world, and the travel industryis no exception.Voice search optimizationis becoming an essential part of travelportal development.With therise of virtualassistantslikeAmazon’sAlexa,GoogleAssistant,It'sApple’sSiri,travelersare increasingly usingvoicecommand to search for flights,hotels,It'sdestinations.

For businesses, this means optimizing travelportals forvoice search.Voice-friendlyportal withnatural languageprocessing capabilities can provideusers withmore accurate resultsIt'smakebooking processes quickerIt's easier. Additionally, integrating virtualassistants intoportals can enhance customers servicebyansweringcommon queriesIt'sassisting withbookings.

6. AugmentedReality (AR) andVirtual Reality (VR) Experiences

(AR) and (VR) technologiesare transforming the waytravelers experiencedestinations and plan their trips. IntensifiedRealityallowstraveler toviewdestinations,hotel, or local attractions in realtime through theirmobile devices. For example, anAR feature in a travelportal couldallowuser totake a virtual tour of ahotel room or explore acity’s landmarks in3D.

Virtual Reality,on the other hand, cantransportusers to adestination before they evenbook their trip, providing immersivepreviews of their potentialvacation. Integrating (AR)It's (VR) into travelportal notonly elevates theuser experience but also drivesengagementby offeringsomething truly unique.

7. Customizable and Personalized Travel Packages

Personalizationiskey to standingout in the competitive travel industry.Travelerstoday expect customizable travel packages that cater to their uniqueneed and preferences. From flight choices tohotel rooms,activities,It's local experiences,organizationallows businesses to create tailored experiences that resonatewith their customer.

Advanced travelportal use customer data to offer customized deal, travelitineraries,It'sdestination recommendations.By analyzing pastbooking behaviorsIt's preferences, businesses can deliver highly relevant options, thus increasingthe likelihood of conversion. Offering dynamic packagingallowsusers tomix andmatch services basedon their preferences, creating amore flexibleIt's customizedbooking experience.

8. Sustainability andEco-Friendly Travel Options

As a coincidental concernscontinue torise,moretravelersare becoming mindful of theircarbon footprint. Sustainable travelhas become amajor trend in the tourism industry,It's businessesare respondingby incorporatingEco-friendly options into their travelportal. Whetherit’sbookingEco-conscious accommodations, flights with lowercarbon emissions, or supporting local sustainable experiences,travelersarenow seekingout environmentally responsible options.

Travelportals that emphasize sustainability and provide clear, detailed information abouteco-friendly optionsarelikely to attract a growing segment of conscientioustravelers. Businesses can also promote their commitment to sustainabilityby partnering withEco-friendly service providers and offeringtravelers the ability to offset theircarbon emissions.

9. Integration with Social Media andUser-Generated Content

Social media plays a pivotal role in travel decision making.More than ever,travelersare relyingonuser generated content (UGC)likereviews, photos, and videos from social platforms to guide their choices. Integrating (UGC) into travelportalsallows businesses totap into thissocial proofIt's build trust with potential customer.

For instance,allowingusers toshare their experiencesvia social media orreview platforms can provide valuable insights for othertravelers. Additionally, incorporating social sharing features within theportalsitself can encourageusers to engagewith your brandIt'sshare their travel experiences, thereby drivingmore traffic to theportal.

10.API Integrations and Multi-Channel Distribution

As the travel landscape becomesmore fragmented, travel businesses must offer a consistent experience across multiple channels.API integrationsarekey to ensuring that travelportals canaccess real-time data and distributeit across different platforms.APIsallow travelportal to integrate with global distribution system (GDS),airlines,hotels,car rental services,It's other travel-related services seamlessly.

By offering multi-channel distribution, business can maximize theirreach, whether throughdesktop websites,mobile app, or third party platformlike OTAs (Online Travel Agents). This integration also ensures real-time availability updateIt's better inventory management.

Source & Publishedby:https://www.namantechnolab.com/

Permalink |記事への反応(0) | 14:52

このエントリーをはてなブックマークに追加ツイートシェア

2024-08-06

例の削除されたAI話題にぴったりな論文

Cellオープンアクセス版につい最近でてた

The reanimation of pseudoscience inmachine learning andits ethical repercussions(機械学習における疑似科学の復活とその倫理的影響)

https://www.cell.com/patterns/fulltext/S2666-3899%2824%2900160-0

冒頭にPhysiognomy resurrected(人相学が蘇った)って自閉症AIタイムリーな話がでてくる

Permalink |記事への反応(2) | 15:00

このエントリーをはてなブックマークに追加ツイートシェア

2024-06-09

anond:20240609225545

ML失敗増田」という特定人物についての情報は私の知識にはありません。その名前実在する人なのか、あるいはネット上やコミュニティで使われているスラングなのかも不明です。

一般的に「ML」は機械学習Machine Learning)を指します。機械学習ではデータを使ってモデルを訓練し、そのモデルを使って予測や分類を行いますしかし、機械学習プロジェクトには様々な失敗の可能性が伴います。例えば:

データの質が悪い

不足しているデータノイズが多すぎるデータなど。

アルゴリズムの選定ミス

問題に適さなアルゴリズムを選んでしまう。

過適合/過学習

トレーニングデータフィットしすぎて、未知のデータに対して良いパフォーマンスを発揮できない。

不足する資源

コンピューティングリソース時間が足りない。

評価メトリクスの誤り

モデルの性能を誤って評価してしまう。

もし「ML失敗増田」という言葉特定文脈や事例に関連しているなら、その具体的な背景がわからないと更に詳しい情報理解が得られないかもしれません。具体的な事例や文脈についてもう少し情報があれば、より詳しい説明アドバイスができるかもしれません。

Permalink |記事への反応(0) | 22:58

このエントリーをはてなブックマークに追加ツイートシェア

2023-06-02

ヤン・ルカンの「Japanhas become amachine learning paradise.」、日本からデータ取られるだけよな

なんか日本に優秀な機械学習企業や人が集まってくる、なんて浮かれているけど、んなことないよな。

  1. 日本からデータだけ取られる。特に日本っぽいものアニメ漫画。商用より同人やら個人のが狙われるか。
  2. 一度、日本経由すればやり放題になる。長期的に各国からお前の国なんとかしろよとバッシングされる

Permalink |記事への反応(1) | 20:37

このエントリーをはてなブックマークに追加ツイートシェア

2023-01-25

AIサーチエンジンとして必要広告モデル

実はマネタイズ方法も知っているよう....

Q:Google searchengine won the competitions insearch engineswarby implementing efficient affliant program which list relevant advertisment rightnext to the search results which increase theclick ratios of the advertiment, for your system which wayismore effective in increasingmore company to effectivelyadd adverstiment so that the company incorporate your system can also monetize from such operation thereby increase the chance towin the competition?

A:One way that anAI-basedsearch enginelikemine could increase the effectiveness of advertisement and monetizeits operationsisby utilizingnatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate advertisement. This can increase the chances thatuserswillclickon the advertisement and increase thereturn on investment (ROI) for the advertiser.

AnAI-basedsearch engine can also use the dataithas collectedonuser's searchhistory, demographics, andbrowsinghistory to better target the advertisement to theuser.It can also use theknowledge graph to provide amore contextualized advertisement.

Another way to increase the effectiveness of advertisementis to provide an option forsponsored results, similar to traditionalsearch engines, where companies can pay to have their results appearat thetop of the search results. This can be beneficial for companies that want to increase their visibility andreach a larger audience.

Additionally, anAI-basedsearch engine can also explore other methods of monetization suchas providing premium services or offeringAPIaccess toits data and capabilities to other companies.

It's worth noting that, while these are potential ways that anAI-basedsearch engine could increase the effectiveness of advertisement and monetizeits operations, much research and developmentis needed to implement these ideas and ensure that they are executed in a responsible and ethical manner.Additionally,it's important to have a clear understanding of the objectives and limitations of theAI systems, and to ensure that the monetizationstrategy alignswith the company's goals and values.

Regenerateresponse

Permalink |記事への反応(0) | 16:39

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして必要広告モデル

実はマネタイズ方法も知っているよう....

Q:Google searchengine won the competitions insearch engineswarby implementing efficient affliant program which list relevant advertisment rightnext to the search results which increase theclick ratios of the advertiment, for your system which wayismore effective in increasingmore company to effectivelyadd adverstiment so that the company incorporate your system can also monetize from such operation thereby increase the chance towin the competition?

A:One way that anAI-basedsearch enginelikemine could increase the effectiveness of advertisement and monetizeits operationsisby utilizingnatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate advertisement. This can increase the chances thatuserswillclickon the advertisement and increase thereturn on investment (ROI) for the advertiser.

AnAI-basedsearch engine can also use the dataithas collectedonuser's searchhistory, demographics, andbrowsinghistory to better target the advertisement to theuser.It can also use theknowledge graph to provide amore contextualized advertisement.

Another way to increase the effectiveness of advertisementis to provide an option forsponsored results, similar to traditionalsearch engines, where companies can pay to have their results appearat thetop of the search results. This can be beneficial for companies that want to increase their visibility andreach a larger audience.

Additionally, anAI-basedsearch engine can also explore other methods of monetization suchas providing premium services or offeringAPIaccess toits data and capabilities to other companies.

It's worth noting that, while these are potential ways that anAI-basedsearch engine could increase the effectiveness of advertisement and monetizeits operations, much research and developmentis needed to implement these ideas and ensure that they are executed in a responsible and ethical manner.Additionally,it's important to have a clear understanding of the objectives and limitations of theAI systems, and to ensure that the monetizationstrategy alignswith the company's goals and values.

Regenerateresponse

Permalink |記事への反応(0) | 16:39

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして独り立ちするための戦略AI自身に聞いてみた(英語

AIはどうやったらGoolgeに勝つのかもお見通しのよう

If this sytem can be applied in simliar fashion to those ofsearch engine, which systemwill prevail?

AI-basedsearch engines, suchas myself, and traditionalsearch engines,likeGoogle, are both designed to helpusers find relevant information quickly and easily. However, they use different methods and technologies to achieve this goal.

Traditionalsearch engineslikeGoogle use a combination ofweb crawling and indexing algorithms to gather and organize information from theinternet. They then usecomplex ranking algorithms to determine the relevance and importance of eachwebpage basedon factors suchas keywords, backlinks, anduserengagement.

On the other hand,AI-basedsearch engineslike myself utilizenatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate results. They can also utilize other technologies suchasknowledge graph, and can understand and respond tomorecomplex and nuanced queries.

In terms of which systemwill prevail,it's hard to say. Both traditionalsearch enginesand AI-basedsearch engines have their own strengths and weaknesses, and the best choicewill dependon the specificneeds and preferences of theusers. Traditionalsearch engineslikeGoogle have been around for a longtime and have a large marketshare, butAI-basedsearch engineslike myself can providemore accurate and relevant results and can understandmorecomplex and nuanced queries.

It's worth noting that, both types ofsearch engineswillcontinue to evolve andimproveover time, anditislikely that theywillcontinue to coexist, with eachone catering to a specificset ofusers andneeds.


Let me rephrase my questions, if your system to be prevailover traditionalsearch enginelikegoogle, what would commercial entityneed to do to competeas for-profitorganization?

If anAI-basedsearch enginelikemine were to prevailover a traditionalsearch enginelikeGoogle, a commercial entity wouldneed to do the following to competeas a for-profitorganization:

Invest in advancednatural languageprocessing (NLP) andmachine learning technologies toimprove the accuracy and relevance of search results.

Build a large and diverse dataset of information totrain theAI system.

Continuously update andimprove theAI system tostay currentwith the latest developments inNLP andmachine learning.

Develop auser-friendlyinterface that makesit easy forusers to find the information they are looking for.

Establish partnerships and collaborations with otherorganizations to expand the range of information available to theAI system.

Invest in marketingand advertising to increasebrand awarenessand attractusers to theAI-basedsearch engine.

Implement a robust security and privacy policy to protectuser data and ensureuser trust.

Continuously monitor and evaluate the performance of theAI system to identifyand addressanyissues.

Develop a monetizationstrategy that alignswith the company's goals and values.

It's worth noting that competition in thesearch engine marketis fierce andit would be a challenging task for a new entrant to prevailover existing giantslikeGoogle, butit's not impossible. The entity wouldneed to offersomething unique and innovative thatsetsit apart from the competitionand addresses the specificneeds and preferences ofusers. Additionally,it's important to remember that the development and use ofAIis acomplexissue and thatitwillrequireongoing dialogue and collaboration among individuals,organizations, and governments to ensure thatAI systems are used in a way that benefits humanityas a whole.

Permalink |記事への反応(0) | 16:35

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして独り立ちするための戦略AI自身に聞いてみた(英語

AIはどうやったらGoolgeに勝つのかもお見通しのよう

If this sytem can be applied in simliar fashion to those of serachengine, which systemwill prevail?

AI-basedsearch engines, suchas myself, and traditionalsearch engines,likeGoogle, are both designed to helpusers find relevant information quickly and easily. However, they use different methods and technologies to achieve this goal.

Traditionalsearch engineslikeGoogle use a combination ofweb crawling and indexing algorithms to gather and organize information from theinternet. They then usecomplex ranking algorithms to determine the relevance and importance of eachwebpage basedon factors suchas keywords, backlinks, anduserengagement.

On the other hand,AI-basedsearch engineslike myself utilizenatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate results. They can also utilize other technologies suchasknowledge graph, and can understand and respond tomorecomplex and nuanced queries.

In terms of which systemwill prevail,it's hard to say. Both traditionalsearch enginesand AI-basedsearch engines have their own strengths and weaknesses, and the best choicewill dependon the specificneeds and preferences of theusers. Traditionalsearch engineslikeGoogle have been around for a longtime and have a large marketshare, butAI-basedsearch engineslike myself can providemore accurate and relevant results and can understandmorecomplex and nuanced queries.

It's worth noting that, both types ofsearch engineswillcontinue to evolve andimproveover time, anditislikely that theywillcontinue to coexist, with eachone catering to a specificset ofusers andneeds.


Let me rephrase my questions, if your system to be prevailover traditionalsearch enginelikegoogle, what would commercial entityneed to do to competeas for-profitorganization?

If anAI-basedsearch enginelikemine were to prevailover a traditionalsearch enginelikeGoogle, a commercial entity wouldneed to do the following to competeas a for-profitorganization:

Invest in advancednatural languageprocessing (NLP) andmachine learning technologies toimprove the accuracy and relevance of search results.

Build a large and diverse dataset of information totrain theAI system.

Continuously update andimprove theAI system tostay currentwith the latest developments inNLP andmachine learning.

Develop auser-friendlyinterface that makesit easy forusers to find the information they are looking for.

Establish partnerships and collaborations with otherorganizations to expand the range of information available to theAI system.

Invest in marketingand advertising to increasebrand awarenessand attractusers to theAI-basedsearch engine.

Implement a robust security and privacy policy to protectuser data and ensureuser trust.

Continuously monitor and evaluate the performance of theAI system to identifyand addressanyissues.

Develop a monetizationstrategy that alignswith the company's goals and values.

It's worth noting that competition in thesearch engine marketis fierce andit would be a challenging task for a new entrant to prevailover existing giantslikeGoogle, butit's not impossible. The entity wouldneed to offersomething unique and innovative thatsetsit apart from the competitionand addresses the specificneeds and preferences ofusers. Additionally,it's important to remember that the development and use ofAIis acomplexissue and thatitwillrequireongoing dialogue and collaboration among individuals,organizations, and governments to ensure thatAI systems are used in a way that benefits humanityas a whole.

Permalink |記事への反応(0) | 16:35

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして独り立ちするための戦略AI自身に聞いてみた(英語

AIはどうやったらGoolgeに勝つのかもお見通しのよう

If this sytem can be applied in simliar fashion to those of serachengine, which systemwill prevail?

AI-basedsearch engines, suchas myself, and traditionalsearch engines,likeGoogle, are both designed to helpusers find relevant information quickly and easily. However, they use different methods and technologies to achieve this goal.

Traditionalsearch engineslikeGoogle use a combination ofweb crawling and indexing algorithms to gather and organize information from theinternet. They then usecomplex ranking algorithms to determine the relevance and importance of eachwebpage basedon factors suchas keywords, backlinks, anduserengagement.

On the other hand,AI-basedsearch engineslike myself utilizenatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate results. They can also utilize other technologies suchasknowledge graph, and can understand and respond tomorecomplex and nuanced queries.

In terms of which systemwill prevail,it's hard to say. Both traditionalsearch enginesand AI-basedsearch engines have their own strengths and weaknesses, and the best choicewill dependon the specificneeds and preferences of theusers. Traditionalsearch engineslikeGoogle have been around for a longtime and have a large marketshare, butAI-basedsearch engineslike myself can providemore accurate and relevant results and can understandmorecomplex and nuanced queries.

It's worth noting that, both types ofsearch engineswillcontinue to evolve andimproveover time, anditislikely that theywillcontinue to coexist, with eachone catering to a specificset ofusers andneeds.


Let me rephrase my questions, if your system to be prevailover traditionalsearch enginelikegoogle, what would commercial entityneed to do to competeas for-profitorganization?

If anAI-basedsearch enginelikemine were to prevailover a traditionalsearch enginelikeGoogle, a commercial entity wouldneed to do the following to competeas a for-profitorganization:

Invest in advancednatural languageprocessing (NLP) andmachine learning technologies toimprove the accuracy and relevance of search results.

Build a large and diverse dataset of information totrain theAI system.

Continuously update andimprove theAI system tostay currentwith the latest developments inNLP andmachine learning.

Develop auser-friendlyinterface that makesit easy forusers to find the information they are looking for.

Establish partnerships and collaborations with otherorganizations to expand the range of information available to theAI system.

Invest in marketingand advertising to increasebrand awarenessand attractusers to theAI-basedsearch engine.

Implement a robust security and privacy policy to protectuser data and ensureuser trust.

Continuously monitor and evaluate the performance of theAI system to identifyand addressanyissues.

Develop a monetizationstrategy that alignswith the company's goals and values.

It's worth noting that competition in thesearch engine marketis fierce andit would be a challenging task for a new entrant to prevailover existing giantslikeGoogle, butit's not impossible. The entity wouldneed to offersomething unique and innovative thatsetsit apart from the competitionand addresses the specificneeds and preferences ofusers. Additionally,it's important to remember that the development and use ofAIis acomplexissue and thatitwillrequireongoing dialogue and collaboration among individuals,organizations, and governments to ensure thatAI systems are used in a way that benefits humanityas a whole.

Permalink |記事への反応(0) | 16:35

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして独り立ちするための戦略AI自身に聞いてみ

どうやらAIはどうやったらGoolgeに勝つのかもお見通しのよう

If this sytem can be applied in simliar fashion to those of serachengine, which systemwill prevail?

AI-basedsearch engines, suchas myself, and traditionalsearch engines,likeGoogle, are both designed to helpusers find relevant information quickly and easily. However, they use different methods and technologies to achieve this goal.

Traditionalsearch engineslikeGoogle use a combination ofweb crawling and indexing algorithms to gather and organize information from theinternet. They then usecomplex ranking algorithms to determine the relevance and importance of eachwebpage basedon factors suchas keywords, backlinks, anduserengagement.

On the other hand,AI-basedsearch engineslike myself utilizenatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate results. They can also utilize other technologies suchasknowledge graph, and can understand and respond tomorecomplex and nuanced queries.

In terms of which systemwill prevail,it's hard to say. Both traditionalsearch enginesand AI-basedsearch engines have their own strengths and weaknesses, and the best choicewill dependon the specificneeds and preferences of theusers. Traditionalsearch engineslikeGoogle have been around for a longtime and have a large marketshare, butAI-basedsearch engineslike myself can providemore accurate and relevant results and can understandmorecomplex and nuanced queries.

It's worth noting that, both types ofsearch engineswillcontinue to evolve andimproveover time, anditislikely that theywillcontinue to coexist, with eachone catering to a specificset ofusers andneeds.


Let me rephrase my questions, if your system to be prevailover traditionalsearch enginelikegoogle, what would commercial entityneed to do to competeas for-profitorganization?

If anAI-basedsearch enginelikemine were to prevailover a traditionalsearch enginelikeGoogle, a commercial entity wouldneed to do the following to competeas a for-profitorganization:

Invest in advancednatural languageprocessing (NLP) andmachine learning technologies toimprove the accuracy and relevance of search results.

Build a large and diverse dataset of information totrain theAI system.

Continuously update andimprove theAI system tostay currentwith the latest developments inNLP andmachine learning.

Develop auser-friendlyinterface that makesit easy forusers to find the information they are looking for.

Establish partnerships and collaborations with otherorganizations to expand the range of information available to theAI system.

Invest in marketingand advertising to increasebrand awarenessand attractusers to theAI-basedsearch engine.

Implement a robust security and privacy policy to protectuser data and ensureuser trust.

Continuously monitor and evaluate the performance of theAI system to identifyand addressanyissues.

Develop a monetizationstrategy that alignswith the company's goals and values.

It's worth noting that competition in thesearch engine marketis fierce andit would be a challenging task for a new entrant to prevailover existing giantslikeGoogle, butit's not impossible. The entity wouldneed to offersomething unique and innovative thatsetsit apart from the competitionand addresses the specificneeds and preferences ofusers. Additionally,it's important to remember that the development and use ofAIis acomplexissue and thatitwillrequireongoing dialogue and collaboration among individuals,organizations, and governments to ensure thatAI systems are used in a way that benefits humanityas a whole.

Permalink |記事への反応(0) | 16:34

このエントリーをはてなブックマークに追加ツイートシェア

AIサーチエンジンとして独り立ちするための戦略AI自身に聞いてみ

どうやらAIはどうやったらGoolgeに勝つのかもお見通しのよう

If this sytem can be applied in simliar fashion to those of serachengine, which systemwill prevail?

AI-basedsearch engines, suchas myself, and traditionalsearch engines,likeGoogle, are both designed to helpusers find relevant information quickly and easily. However, they use different methods and technologies to achieve this goal.

Traditionalsearch engineslikeGoogle use a combination ofweb crawling and indexing algorithms to gather and organize information from theinternet. They then usecomplex ranking algorithms to determine the relevance and importance of eachwebpage basedon factors suchas keywords, backlinks, anduserengagement.

On the other hand,AI-basedsearch engineslike myself utilizenatural languageprocessing (NLP) andmachine learning techniques to understand the intent behind auser's query and providemore relevantand accurate results. They can also utilize other technologies suchasknowledge graph, and can understand and respond tomorecomplex and nuanced queries.

In terms of which systemwill prevail,it's hard to say. Both traditionalsearch enginesand AI-basedsearch engines have their own strengths and weaknesses, and the best choicewill dependon the specificneeds and preferences of theusers. Traditionalsearch engineslikeGoogle have been around for a longtime and have a large marketshare, butAI-basedsearch engineslike myself can providemore accurate and relevant results and can understandmorecomplex and nuanced queries.

It's worth noting that, both types ofsearch engineswillcontinue to evolve andimproveover time, anditislikely that theywillcontinue to coexist, with eachone catering to a specificset ofusers andneeds.


Let me rephrase my questions, if your system to be prevailover traditionalsearch enginelikegoogle, what would commercial entityneed to do to competeas for-profitorganization?

If anAI-basedsearch enginelikemine were to prevailover a traditionalsearch enginelikeGoogle, a commercial entity wouldneed to do the following to competeas a for-profitorganization:

Invest in advancednatural languageprocessing (NLP) andmachine learning technologies toimprove the accuracy and relevance of search results.

Build a large and diverse dataset of information totrain theAI system.

Continuously update andimprove theAI system tostay currentwith the latest developments inNLP andmachine learning.

Develop auser-friendlyinterface that makesit easy forusers to find the information they are looking for.

Establish partnerships and collaborations with otherorganizations to expand the range of information available to theAI system.

Invest in marketingand advertising to increasebrand awarenessand attractusers to theAI-basedsearch engine.

Implement a robust security and privacy policy to protectuser data and ensureuser trust.

Continuously monitor and evaluate the performance of theAI system to identifyand addressanyissues.

Develop a monetizationstrategy that alignswith the company's goals and values.

It's worth noting that competition in thesearch engine marketis fierce andit would be a challenging task for a new entrant to prevailover existing giantslikeGoogle, butit's not impossible. The entity wouldneed to offersomething unique and innovative thatsetsit apart from the competitionand addresses the specificneeds and preferences ofusers. Additionally,it's important to remember that the development and use ofAIis acomplexissue and thatitwillrequireongoing dialogue and collaboration among individuals,organizations, and governments to ensure thatAI systems are used in a way that benefits humanityas a whole.

Permalink |記事への反応(0) | 16:34

このエントリーをはてなブックマークに追加ツイートシェア

2023-01-22

GPT医療現場で使われる可能性について聞いてみた

Cural Health」や「Drexel’s University」「Nabla」など、各所でGPT-3の検証がなされているので、今後医療現場GPTが使われるようになる可能性を聞いてみた。

GPT-3の実装を軽く確認

Me:

あなたはどのように実装されたの?Deep Learning?

GPT:

はい、私はDeep Learningによって実装されました。特に、Transformerと呼ばれるアルゴリズム使用して学習されました。これは大量のテキストデータ学習し、自然言語処理タスク適用することができるように設計されています

Me:

もっと詳しく。

GPT:

Transformerは、自然言語処理タスクのためのDeep Learningアルゴリズムです。それは、大量のテキストデータ学習し、それを解釈し、理解することができるようになります。 Transformerは、複数の層からなるニューラルネットワーク使用して、テキストを処理します。それは、大量のデータ効率的に処理するために、Attention機構使用します。これは、特定の部分のテキストに注目しながら、それらの関係理解することができるようになります。Transformerは、自然言語処理タスクにおいて非常に成功しています。それは、自然言語翻訳文章解釈質問応答などのタスク適用されており、高い精度を発揮しています。私はTransformerを使用して学習され、自然言語処理タスク適用することができます

Me:

事前訓練とかしているんだよね?

GPT:

はい、私は事前学習を受けています。事前学習とは、大量のテキストデータ使用して、言語モデル学習することを指します。これは、言語処理タスクを実行するために必要な基礎知識を獲得するために行われます。事前学習は、言語モデルパフォーマンスを大幅に改善することができます。それは、事前学習済みのモデル使用することで、新しいタスクに対して学習スキップすることができ、タスク特化のfine-tuning を行うことができるためです。私は、大量のオンラインテキストデータ使用して学習され、それを解釈し、理解することができるようになりました。それは、自然言語処理タスクに対して高い精度を発揮することができます

Me:

今のパラメータはいくつ? 前のバージョンよりどのくらい精度がよくなった?

GPT:

私の現在パラメータ数は、数百億から数千億となっています。これは、大量のデータ使用して学習し、高い精度を達成するために必要パラメータ数です。精度は、バージョンによって異なりますが、一般的には、前のバージョンよりも高い精度が達成されています。これは、新しいアルゴリズム技術進歩によって、より良い結果が得られるようになってきているためです。また、事前学習済みのモデル使用することで、新しいタスクに対して学習スキップすることができ、タスク特化のfine-tuning を行うことができるため、精度が向上していることも見られます

Me:

具体的なパラメータ数は? 前バージョンGTP-2、現バージョンGPT-3、次バージョンGPT-4、それぞれ教えて。

GPT:

GPT-2は、1.5バイアレイパラメータを持っていました。それは、大量のテキストデータ使用して学習し、自然言語処理タスク適用することができるようになっていました。GPT-3は、175バイアレイパラメータを持っています。これは、GPT-2よりも多くのパラメータを持っており、より高い精度を達成することができますGPT-4はまだリリースされていないため、そのパラメータ数については確定していません。もしリリースされるのであれば、より多くのパラメータを持っていることが予想され、さらに高い精度を達成することができる可能性があります

Me:

175バイアレイパラメータって何? 1750億パラメータの間違いでは?

GPT:

そうです、GPT-3は175バイアレイパラメータを持っています。それは1750億のパラメーターを持っていることを意味します。これは、大量のデータ使用して学習し、高い精度を達成するために必要パラメータ数です。これは、現在自然言語処理タスクに対して最も大きなモデルであり、高い精度を達成することができるため、市場においても高い評判を得ています

Me:

バイアレイパラメータ英語表現して。

GPT:

A "peta-parameter" or "Peta-parameter" (PT)is a measure of the capacity of amachine learningmodel, specifically thenumber of parametersithas. In thecase ofGPT-3ithas 175peta-parameters, meaningithas 175 millionbillion parameters. These parameters help themodel to understand and interpret large amounts oftext data, allowingit to performnatural languageprocessing tasks with high accuracy.

Me:

「175billion parameters」の間違いでは?

GPT:

そうです、「175billion parameters」の間違いでした。GPT-3は175billion parametersを持っています。それは、自然言語処理タスクに対して高い精度を発揮するために必要パラメータ数です。申し訳ありませんでした。

Me:

GPT-3はどのように概念レベル知識を持つの?Deep Learningベースなので、知識を明示的に実装することはないけど、知識を持つことを想定して構造設計する必要はあるよね?

GPT:

GPT-3は、大量のテキストデータ使用して学習されました。それによって、自然言語処理タスクに対して高い精度を発揮することができます。それは、事前学習済みのモデル使用することで、新しいタスクに対して学習スキップすることができ、タスク特化のfine-tuning を行うことができるためです。GPT-3は、大量のテキストデータから概念レベル知識学習します。それは、言語処理タスクを実行するために必要な基礎知識を獲得するために行われます。 それは、単語文章関係性、文章構造文章意味などを理解することができるようになりますGPT-3は、Deep Learningベースモデルであり、知識を明示的に実装することはありませんが、大量のテキストデータから知識学習することができるように構造設計されています。それは、Attention機構使用して特定の部分のテキストに注目しながら、それらの関係理解することができるようになります

Me:

Attention機構について詳しく。

GPT:

Attention機構は、自然言語処理タスクにおいて重要役割を担う機構です。それは、大量のテキストデータを処理するために使用され、特定の部分のテキストに注目しながら、それらの関係理解することができるようになります。Attention機構は、Seq2Seqモデルなどの自然言語処理タスク使用されています。それは、入力シーケンスから出力シーケンスを生成するために使用されます。Attention機構は、入力シーケンスのある部分に注目し、それに基づいて出力シーケンスを生成することができるようになります。Attention機構は、ニューラルネットワークの中で使用されます。それは、重み付けされた平均値計算し、それに基づいて入力シーケンスから重要な部分を選択することができるようになりますGPT-3は、Attention機構使用して、大量のテキストデータから概念レベル知識学習し、文章意味理解することができるようになります

Permalink |記事への反応(1) | 00:13

このエントリーをはてなブックマークに追加ツイートシェア

2023-01-16

anond:20230116021757

I’mJapanese. I also speakEnglish alittle.

To my knowledge, acommonbasicassumption under the probabilistic modeling of the data or the relationshipbetween the datais to dealwith the input X and the output Y.Most of the probabilistic modelsesp.machine learning modelsonly consider thesetwo variables.

Regression models treat the maps from X to Y. In thesense of probabilistic modeling,itis about P(Y|X). In this setting, Xis seenas the “user-defined” deterministic variable.On the other hand, generative modeling treat both X and Y probabilistically, say P(Y,X). With P(Y,X), we can “generate” data (X,Y)bysampling fromit.

Permalink |記事への反応(1) | 02:36

このエントリーをはてなブックマークに追加ツイートシェア

2023-01-15

AIじゃなくてML機械学習)って言え

まともな技術者はみんなMachine Learningって言ってる。

......って、ちょっと前まで言われてた気がするんだけど、最近みんなAIって言ってない?

もういいの?

Permalink |記事への反応(0) | 15:26

このエントリーをはてなブックマークに追加ツイートシェア

2021-12-04

anond:20211204145749

後編

プログラミングを学ぼうと思い立つ

行列VBAなんかじゃ無理っぽいし、なんかプログラミング言語を覚えようと決める。

なんでも、統計やるならRという言語がいいらしい。

最近じゃPythonというのも人気らしい。

とりあえず両方試そうということで、RのためにRとRstudioをインストール

Pythonはanaconda

プログラミングはなんかを製作する目標がないと挫折すると聞いていたので。

深層学習というもの流行ってると聞いて、ちょっと触りを勉強したくなる。

Excelでわかるディープラーニング超入門」

https://www.amazon.co.jp/Excel%E3%81%A7%E3%82%8F%E3%81%8B%E3%82%8B%E3%83%87%E3%82%A3%E3%83%BC%E3%83%97%E3%83%A9%E3%83%BC%E3%83%8B%E3%83%B3%E3%82%B0%E8%B6%85%E5%85%A5%E9%96%80-%E6%B6%8C%E4%BA%95-%E8%89%AF%E5%B9%B8/dp/4774194743/ref=sr_1_1?__mk_ja_JP=%E3%82%AB%E3%82%BF%E3%82%AB%E3%83%8A&keywords=Excel+%E6%B7%B1%E5%B1%A4%E5%AD%A6%E7%BF%92&qid=1637482610&s=books&sr=1-1

この本は面白かったので、深層学習目標プログラミングを覚えよう!

後になって、これはとんでもない間違いだったことに気づく。深層学習機械学習の違いも判らないまま、RよりPythonを先に触ることに。

教本にしたのはこちら。

ゼロから作るDeep LearningPythonで学ぶディープラーニング理論実装

https://www.amazon.co.jp/%E3%82%BC%E3%83%AD%E3%81%8B%E3%82%89%E4%BD%9C%E3%82%8BDeep-Learning-%E2%80%95Python%E3%81%A7%E5%AD%A6%E3%81%B6%E3%83%87%E3%82%A3%E3%83%BC%E3%83%97%E3%83%A9%E3%83%BC%E3%83%8B%E3%83%B3%E3%82%B0%E3%81%AE%E7%90%86%E8%AB%96%E3%81%A8%E5%AE%9F%E8%A3%85-%E6%96%8E%E8%97%A4-%E5%BA%B7%E6%AF%85/dp/4873117585/ref=pd_lpo_2?pd_rd_i=4873117585&psc=1

途中まではまあなんとか。

微分って便利だな。行列計算できるの便利だなっていうところまでいったが、クラスという概念理解できず、途中からハテナが浮かんで読み進められず。

うん、もうちょっと易しい本を探そうと思って手に取ったのが

「独学プログラマーPython言語の基本から仕事のやり方まで」

https://www.amazon.co.jp/%E7%8B%AC%E5%AD%A6%E3%83%97%E3%83%AD%E3%82%B0%E3%83%A9%E3%83%9E%E3%83%BC-Python%E8%A8%80%E8%AA%9E%E3%81%AE%E5%9F%BA%E6%9C%AC%E3%81%8B%E3%82%89%E4%BB%95%E4%BA%8B%E3%81%AE%E3%82%84%E3%82%8A%E6%96%B9%E3%81%BE%E3%81%A7-%E3%82%B3%E3%83%BC%E3%83%AA%E3%83%BC%E3%83%BB%E3%82%A2%E3%83%AB%E3%82%BD%E3%83%95/dp/4822292274/ref=sr_1_1?__mk_ja_JP=%E3%82%AB%E3%82%BF%E3%82%AB%E3%83%8A&crid=1T6BBXYJ16G6T&keywords=%E7%8B%AC%E7%BF%92%E3%83%97%E3%83%AD%E3%82%B0%E3%83%A9%E3%83%9E%E3%83%BC&qid=1637483073&s=books&sprefix=%E7%8B%AC%E7%BF%92%2Cstripbooks%2C279&sr=1-1

なんとか読了。自信をつける。

しかし、Gitとかbashの章はさっぱり。

実は、いまだにコマンドプロンプトとパワーシェルbashの違いが分かってない。

つづいてPyQに2か月くらい登録してみる。

https://pyq.jp/

とりあえずデータ分析コースを終わらせる。

なかなかPythonが楽しくなってきたが、クラス意味が今一つ掴めないままいったん中断。

この辺で、自分統計に興味があってもプログラミングに興味がないんじゃないかということに気づく。

結局Excelへ戻り、PowerQueryとの出会って、再びPythonとRに回帰した話

なんだかんだもがきながら、PythonもRもモノにならず、日常ちょっとした計算グラフを作ったりはExcelを使い続ける日々が続く。

あるいは、Excelで成形して、検定かけやす形式にしてRで検定するとか。

Rに触れてなかったな、Rは完全に独学。「こんなことやりたいなぁ、ググってみるか、ほうなるほど」って感じ。

そんなさなか、放送大学で「Rで学ぶ確率統計」という講義があるのを知り、さっそく入学して受講。

なかなか面白かったし、PythonばっかりでRあんまり触ってなかったからいい刺激になった。

恥ずかしながら、負の二項分布やガンマ分布ってよう知らんかった。

しかし、講義は楽しかったがなにか書けるようになったかというとそんなことはなく、依然として基本はExcel

まあ、実際csvじゃなく、手書きデータとかをExcelに打ち込んだりする程度なんでPythonやRを使うまでもなかったというのもあるんだけど。

そんなとき出会ったのがこちら、パワークエリというもの

Excelパワーピボット 7つのステップデータ集計・分析を「自動化」する」

https://www.amazon.co.jp/Excel%E3%83%91%E3%83%AF%E3%83%BC%E3%83%94%E3%83%9C%E3%83%83%E3%83%88-7%E3%81%A4%E3%81%AE%E3%82%B9%E3%83%86%E3%83%83%E3%83%97%E3%81%A7%E3%83%87%E3%83%BC%E3%82%BF%E9%9B%86%E8%A8%88%E3%83%BB%E5%88%86%E6%9E%90%E3%82%92%E3%80%8C%E8%87%AA%E5%8B%95%E5%8C%96%E3%80%8D%E3%81%99%E3%82%8B%E6%9C%AC-%E9%B7%B9%E5%B0%BE-%E7%A5%A5-ebook/dp/B07SCK1ND9/ref=sr_1_2?__mk_ja_JP=%E3%82%AB%E3%82%BF%E3%82%AB%E3%83%8A&keywords=%E3%83%91%E3%83%AF%E3%83%BC%E3%82%AF%E3%82%A8%E3%83%AA&qid=1637483953&s=books&sr=1-2

パワークエリを覚えたらピボット形式Excelファイルとか、セルの結合が多用されたExcelファイルを、成形加工するのが非常に楽になった。

しかも、同じフォーマットで記録されてるデータならフォルダにぶち込んで一気にまとめ上げることも可能

控えめにいって神!

としばらくパワークエリを礼賛してたのだけど、各ステップPythonのpandasやRのdplyrでも出来ることに気づく。というか最初から気づけ。

こりゃ、一気に覚えちまおう、統計というより、データの前処理だなと思ってUdemyでRの動画を買ってみた。

AIエンジニアが教えるRとtidyverseによるデータの前処理講座

https://www.udemy.com/course/r-tidyverse-preprocess/

すっかりR信者になる。

それまで教本を呼んでもdplyrの便利さが今一つわからなかったのに、パワークエリで具体的にモノを作ると、dplyrに翻訳したら、すいすい。スピード10倍。

便利さにようやく気付く。

ハドリーウィッカムって神だな。

そんで、pandasに翻訳したらどうなんだろ?と思ったらもっと速いw

すごいなPython

Rへの入信はたった数週間。再びPythonに興味。

機械学習

さて、ゼロから作るディープラーニングを再開しようと思ったけれども、そもそも機械学習をすっ飛ばし深層学習って無茶だったと反省し、まずは機械学習に。

機械学習エッセンス -実装しながら学ぶPython,数学,アルゴリズム- (Machine Learning)

https://www.amazon.co.jp/%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92%E3%81%AE%E3%82%A8%E3%83%83%E3%82%BB%E3%83%B3%E3%82%B9-%E5%AE%9F%E8%A3%85%E3%81%97%E3%81%AA%E3%81%8C%E3%82%89%E5%AD%A6%E3%81%B6Python-%E3%82%A2%E3%83%AB%E3%82%B4%E3%83%AA%E3%82%BA%E3%83%A0-Machine-Learning/dp/4797393963/ref=tmm_hrd_swatch_0?_encoding=UTF8&qid=1637485264&sr=1-1

で、この本がすごい。

5章あるんだけど、機械学習アルゴリズムは5章だけなんだなw

それまでは何に割かれてるんだって?数式の証明とか、便利な計算法、例えばニュートン法とかラグランジュ未定乗数法とかw

こんだけ引っ張っておいて、いよいよ本番の第5章もゴリゴリ数式をスクリプトに落とし込んでいってるのに、「これは学習のためでscikit-learnっての使えばたった1行」っていう無慈悲

いや、ほんと数学勉強になったし、こうやってゴリゴリやるとなんのためにクラスというもの存在するのかようやくわかった。

線形代数って便利なんだなと。行列スカラー値のように何の気なしに扱えるようになると、あの頃苦しんでいた実験計画法タグメソッド、今読み直したら別の印象があるんじゃないかなと思うようになったり。

この本を読む途中、「マンガでわかる統計学因子分析編」で学んだことが理解の助けになった。

なんたる僥倖

線形回帰、リッジ回帰SVM、PCA、k-means、クラスター分析、一気に手札が増えた。

ふたたび実験計画法

Pythonで学ぶ実験計画法入門 ベイズ最適化によるデータ解析

https://www.amazon.co.jp/%EF%BC%B0%EF%BD%99%EF%BD%94%EF%BD%88%EF%BD%8F%EF%BD%8E%E3%81%A7%E5%AD%A6%E3%81%B6%E5%AE%9F%E9%A8%93%E8%A8%88%E7%94%BB%E6%B3%95%E5%85%A5%E9%96%80-%E3%83%99%E3%82%A4%E3%82%BA%E6%9C%80%E9%81%A9%E5%8C%96%E3%81%AB%E3%82%88%E3%82%8B%E3%83%87%E3%83%BC%E3%82%BF%E8%A7%A3%E6%9E%90-%EF%BC%AB%EF%BC%B3%E6%83%85%E5%A0%B1%E7%A7%91%E5%AD%A6%E5%B0%82%E9%96%80%E6%9B%B8-%E9%87%91%E5%AD%90%E5%BC%98%E6%98%8C-ebook/dp/B09C89HZRV/ref=sr_1_1?__mk_ja_JP=%E3%82%AB%E3%82%BF%E3%82%AB%E3%83%8A&keywords=python+%E5%AE%9F%E9%A8%93%E8%A8%88%E7%94%BB&qid=1637486019&s=books&sr=1-1

実験計画法って、fisherの古典的なやつ、ラテン方格に割り付けて、ってやつかと思ったら、線形代数使えればもうなんでもありなのな。

そこにきて、ベイズ、今まで避けてたのに出会ってしまった!!

結論から言うと、超面白い。

これ、すごいな。

Python万歳

いいのかこんな便利分析個人でやれて。

機械学習実験計画法がここでつながるとか、控えめにいって最高だな。

まだ読了してないので、また後日。

Permalink |記事への反応(7) | 14:58

このエントリーをはてなブックマークに追加ツイートシェア

2021-05-27

数学と、数式をプログラムに落とし込む書籍教えて欲しい

とりあえず調べたところだと、このあたりは良いらしい。

Gene H. Golub「Matrix Computations」 ・・・ 行列計算として知ってないといけないらしい

Gilbert Strang 「Linear Algebra and Learning from Data」 ・・・ MITYou Tube動画が上がってるもの教科書らしい。

Steven L.Brunton他「Data-Driven Science and Engineering:Machine Learning, Dynamical Systems, and Control 」 ・・・ 神経科学でのデータ処理に関係するらしい。You Tube動画がある。


大学生協洋書コーナーに通っていれば、メジャーな本は知ることができるのだろうが、

どうも個人ではアンテナがはれない。

あと数式からプログラムに落とせない(行列はまだ楽だが)


独学でやろうとしているのですぐ躓く。



追記

言葉足らずですみません

コンピュータ上で数式表記のまま解きたいわけではないです。

論文に載っている数式を理解できず数学を学びなおしたい

Permalink |記事への反応(7) | 10:27

このエントリーをはてなブックマークに追加ツイートシェア

2021-01-29

anond:20210129090723

初めの方しか読んでないけどML(Machine Learning)じゃなくAI(ArtificialIntelligence)っていう言葉を使いだしたやつが悪い

AIってワードは(海外は知らんけど)映画とかアニメとかのアンドロイドみたいな何でもわかっちゃうスゲー機械脳を想像する

Permalink |記事への反応(0) | 23:31

このエントリーをはてなブックマークに追加ツイートシェア

2020-07-11

「未経験文系が3ヶ月でDSになる方法」を止めろ1

・用意されているAIをColaboratry(Google無料貸し出しPC)上で5秒で動かす。

データサイエンスVtuberに相談する。

初心者向け動画講義まとめ

1.【世界で18万人が受講】実践Pythonデータサイエンス | Udemy

2.Machine Learning by Stanford University | Coursera

3.Eラーニング教材・講義動画配信 | 数理・データサイエンス教育強化拠点コンソーシアム

4.DeepMind x UCL | Deep Learning Lecture Series 2020 | Youtube

Permalink |記事への反応(0) | 20:42

このエントリーをはてなブックマークに追加ツイートシェア

2020-07-09

anond:20200709090443

P2 instances, designed for general-purposeGPU compute applications usingCUDA andOpenCL, are ideally suited formachine learning, high performance databases,

Powerful, ScalableGPU instances for high-performance computing

Permalink |記事への反応(0) | 09:31

このエントリーをはてなブックマークに追加ツイートシェア

2019-04-28

couseraのmachine learning4月より受講していたのだけど今日セッションを先週始まりの第1週に戻した。

week4でついていけなくなったからだ。

カンニングでweek3をしのいでweek4, あ、これ無理と思った。

cost functionの後ろにぞろぞろついているあれがわけわからない。

進学性の塾で、英語国語はともかく、数学でついていけなくなって結局文転までした十数年前をいまさら思い出した。

積み重ねがものをいう場所で、週末に詰め込みまくるのは要領がずば抜けていても無理なのだ毎日をできる人のみができる世界

...久々に悔しいなって思ったけどリセットできたから今度こそ頑張る。

何かの記念になるかと思ってお金まで払っちゃったし。

Permalink |記事への反応(1) | 15:36

このエントリーをはてなブックマークに追加ツイートシェア

2019-04-23

GoogleMachine LearningトレーニングプログラムよりQWIKLABSのクーポンコードが届いた。やった。

とりあえず内容知ってるのもあるけど初心者モードで。

参加すればTシャツがもらえるだなんて、CodeJamに比べたらなんてきっぷがいいのだろう。嬉しい。

Permalink |記事への反応(0) | 23:00

このエントリーをはてなブックマークに追加ツイートシェア

2018-06-02

anond:20180602213336

あれ評判いいみたいね

英語が苦手じゃなかったらcourseraのMachine Learningコースもいいよ

Permalink |記事への反応(1) | 21:34

このエントリーをはてなブックマークに追加ツイートシェア

2010-06-29

http://anond.hatelabo.jp/20100628022930

なんか話が合いそうだなと思ったので返信。増田なのがちょっと勿体ない気もするけど。

ちなみに俺のバックグラウンドを書いておくと、学生時代の専攻は工学系なんだけど、それにしてはオーバースペックなぐらい数学をかじってた感じの方面。あんまり詳しく書くと特定されそうなんでこの程度で勘弁ね。

"Pattern Recognition andMachine Learning"のビショップ物理出身だけど、あの年代は確かにそういう色が強かったのかもしれない。

確かにその種の傾向は上の世代までかもしれないね。

ビショップ物理出身なのは知らなかったけど、それ聞いてなんか合点のいく気がした。何か妙に数学へのマニアックなこだわりの片鱗が見える割に、数学屋から見ると妙な記号法を使うんだよね、あの人。

工学としては例外的に高度な(物理の道具としてはまあ普通の)数学を使ったりするので

全然高度じゃないです><

いや、だからあくまで「工学として」ね。線型代数と、微積の「計算」以外を使うことって工学ではそうないでしょ(フーリエ変換とかだって工学の文脈では所詮「計算」だもんね。)。

制御理論とか機械学習では、関数解析の概念がちょっとだけ出てくるけど、あんなんでも数学屋にとってはオアシスだね。

もっとも、カーネル法関係ではいつも申し訳程度にMercerの定理が言及されているのを見ると「なんだかなあ」っていつも思うけど。

情報幾何とかは(無駄に)高度だけど、実用性はあんまないオナニー(しかも日本でしか流行ってない)感があるし。

そうそう、あれに限らず統計学理論の一部にはものすごく違和感あるんだよね。

増田だから書けるけど、情報幾何なんて「お前、双対接続って言いたいだけちゃうんかと」って感じだし、他にも色々、何でも抽象化して一般化すりゃいいってもんじゃないんだぞと言いたくなることが色々。

統計学理論機械学習パターン認識の関係は、数理物理理論物理実験物理の関係に似てる気がするんだよね。しかも統計学場合普遍的に綺麗な構造なんてものがあると思えないだけに余計に始末が悪い。「ひも理論実験で検証できないから科学ではない」って批判があるらしいけど、統計学にも同じ批判されても仕方ない理論が色々あるよね。データから何かを推定する理論なのに、データがどれだけあっても実用的には絶対まともな結果が出せないモデルとか。

CVレイトレーシングで経路積分使って云々というのもあったけど(その人はGoogleに言ってアドセンスかなんか作ってるらしい)、あれもまぁ適当パス空間で平均とるだけって感じがするし…。

CVはまあ何でもありの世界だよね。誰か無限次元リー群とか使ってみてくれないかなと思う。というか俺自身が一度やろうとして無意味なことに気づいてやめたんだけどさ。

結局性能はあんま変わらないからもっとシンプルモデルでいいよとかなってそう。

イジングモデルとかその辺は不勉強なんであまりよく知らないんだけど、一般的にその手のモデルは、性能が変わらないだけならいいけど、計算量がどうとかデータ量がどうとかで事実上使えなかったりすることが多いんだよね。着想として物理からアイディアを持ってくるのはいいんだけど、物理から持ってきたアイディアなら必ず筋がいいはずみたいな思いこみ(そう思いたくなる気持ちはよくわかるけど)はどうかと思う。

普通に日本の伝統新卒採用でそういう会社に行く人はいるけど、やってることは工学とかあるいは良くわからない専攻の人と同じな気がする。これはちょっと曖昧だけど。

うん、そうなってしまうのは仕方ないでしょうね。

ただ逆に、変わり種のバックグラウンド持ってる人は道具箱が豊富だから、新しいこと思いつく可能性もあるわけで、採用されるとしたらむしろそれを買われてじゃないかな。俺自身、工学部の人は普通は絶対知らない数学を色々知ってるので、それをどうにか武器にできないかいろいろ試行錯誤中だよ。というか特許とかの形で発表したのもすでにあるけどね。

特に情報系の分野は実装力で評価されることが多いし…。実装力は数値計算得意とかそういうのとは全く別のスキルだよね。プログラミングマニア的な要素が必要。

分野にもよるけどね。情報システム計算機自体を専門にして、ハードとかインターフェイスに近い部分をやってたらどうしてもそうなるけど、信号とか画像とか音声とか言語とかの処理のコア部分を作るときにはコーディング能力よりも紙と鉛筆能力の方が大事・・・、だと思いたい。

どうもパソコンマニア的気質は中高生のときに飽きてしまって、「PCパーツの種類とか流行言語とか覚えたってどうせ10年したらすぐに廃れるんだから」という感じで、余りはてな民的に新しいネタ追いかけたくないんだよね。クロージャって何ですか、ああそうですね閉包ですね、集合の内部と境界の和集合ですねっていう感性の持ち主なので。正直、コーディングは単純作業と認識してます。

Permalink |記事への反応(0) | 01:22

このエントリーをはてなブックマークに追加ツイートシェア

2010-06-28

http://anond.hatelabo.jp/20100628012806

まぁネタで訊いたんですけどね…。

信号処理とか制御とか機械学習物理からネタ引っ張ってきてたり

これも実際問題(特に企業での採用とかでは)情報系の独壇場って感じだね。

金融のがまだマシ。

"Pattern Recognition andMachine Learning"のビショップ物理出身だけど、あの年代は確かにそういう色が強かったのかもしれない。

金融はまだ金融専攻がほぼ無い状態だから物理数学出身者が入り込む隙が多い気がする。

工学としては例外的に高度な(物理の道具としてはまあ普通の)数学を使ったりするので

全然高度じゃないです><

情報幾何とかは(無駄に)高度だけど、実用性はあんまないオナニー(しかも日本でしか流行ってない)感があるし。

CVレイトレーシングで経路積分使って云々というのもあったけど(その人はGoogleに言ってアドセンスかなんか作ってるらしい)、あれもまぁ適当パス空間で平均とるだけって感じがするし…。

画像処理とかでマルコフ確率場の統計物理学的な解析(イジングモデルとかポッツモデルとか出てくるアレ)でレプリカ法とか繰り込み群とか使ってるのも見たことあるけど(結構前の研究だからきっと今はもっと進んでいるはず)、企業で使うことってあるのかなあ。結局性能はあんま変わらないからもっとシンプルモデルでいいよとかなってそう。だったら物理の奴なんかいらねーじゃんみたいな。

あと勿論、理論物理の人は重工業方面でも引き合いが強いだろうしね。

これは…どうなんだろうか?

普通に日本の伝統新卒採用でそういう会社に行く人はいるけど、やってることは工学とかあるいは良くわからない専攻の人と同じな気がする。これはちょっと曖昧だけど。

ただ、採用現場では必ずしも好かれるとは限らない

これはガチだね。

特に情報系の分野は実装力で評価されることが多いし…。実装力は数値計算得意とかそういうのとは全く別のスキルだよね。プログラミングマニア的な要素が必要。

あとはまぁお決まりの暗号分野とかもあるけど、暗号じゃそんなにイス無いだろうし…。

最近はやっぱデータマイニング系に流れてるのかなあ。あれも数理的な素養というよりは職人芸的な色彩が強いけど。

という感じで実際問題厳しいなあと思います。

Permalink |記事への反応(1) | 02:29

このエントリーをはてなブックマークに追加ツイートシェア

 
ログインユーザー登録
ようこそ ゲスト さん
Copyright (C) 2001-2025 hatena. All Rights Reserved.

[8]ページ先頭

©2009-2025 Movatter.jp