TheQLattice is asoftware library which provides a framework forsymbolic regression inPython. It works onLinux,Windows, andmacOS. The QLattice algorithm is developed by the Danish/Spanish AI research companyAbzu.[1] Since its creation, the QLattice has attracted significant attention, mainly for the inherent explainability of the models it produces.[2][3][4]
At theGECCO conference in Boston, MA in July 2022, the QLattice was announced as the winner of the synthetic track of theSRBench competition.[5]
The QLattice works with data in categorical and numeric format. It allows the user to quickly generate, plot and inspect mathematical formulae that can potentially explain the generating process of the data. It is designed for easy interaction with the researcher, allowing the user to guide the search based on their preexisting knowledge.[2][6]
The QLattice mainly targets scientists, and integrates well with the scientific workflow.[2][6] It has been used in research into many different areas, such as energy consumption in buildings,[3] water potability,[7] heart failure,[8] pre-eclampsia,[4] Alzheimer's disease,[9] hepatocellular carcinoma,[9] and breast cancer.[9]
^Kevin René Broløs; Meera Vieira Machado; Chris Cave; Jaan Kasak; Valdemar Stentoft-Hansen; Victor Galindo Batanero; Tom Jelen; Casper Wilstrup (2021-04-12). "An Approach to Symbolic Regression Using Feyn".arXiv:2104.05417 [cs.LG].
^abWilstrup, Casper; Hedley, Paula L.; Rode, Line; Placing, Sophie; Wøjdemann, Karen R.; Shalmi, Anne-Cathrine; Sundberg, Karin; Christiansen, Michael (2022-06-30). "Symbolic regression analysis of interactions between first trimester maternal serum adipokines in pregnancies which develop pre-eclampsia".medRxiv10.1101/2022.06.29.22277072.
^Riyantoko, Prismahardi Aji; Diyasa, I Gede Susrama Mas (2021-10-28)."F.Q.A.M" Feyn-QLattice Automation Modelling: Python Module of Machine Learning for Data Classification in Water Potability. IEEE. pp. 135–141.doi:10.1109/icimcis53775.2021.9699371.ISBN978-1-6654-2733-3.
