0481488 20240103214946.020171116235959.9 000427151400050 6 s. E cav_un_epca*0477966978-80-7435-678-0289-294Hradec KrálovéUniversity of Hradec Králové2017 data-based learning probabilistic models information theory MDL principle lossless encoding cav_un_auth*0101118 Matematická teorie rozhodování Department of Decision Making Theory MTR MTR Department of Decision Making Theory 75 Jiroušek Radim UTIA-B K Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0353476 25 Krejčová I. CZ http://library.utia.cas.cz/separaty/2017/MTR/jirousek-0481488.pdf cav_un_auth*0353428 GA15-00215S GA ČR CZ The problem of overfitting is studied from the perspective of information theory. In this context, data-based model learning can be viewed as a transformation process, a process transforming the information contained in data into the information represented by a model. The overfitting of a model often occurs when one considers an unnecessarily complex model, which usually means that the considered model contains more information than the original data. Thus, using one of the basic laws of information theory saying that any transformation cannot increase the amount of information, we get the basic restriction laid on models constructed from data: A model is acceptable if it does not contain more information than the input data file. cav_un_auth*0346896 MME 2017. International Conference Mathematical Methods in Economics /35./ 20170913 20170915 Hradec Králové CZ AH 50000 50200 50202 2018 2 PR RVO:67985556 http://hdl.handle.net/11104/0277045 cav_un_auth*0353468 Výsoká škola ekonomická, Fakulta managementu FM VŠE CZ S 3+4 Proceedings Paper Economics|Operations Research Management Science|Mathematics Interdisciplinary Applications|Social Sciences Mathematical Methods 3+4 Proceedings Paper Economics|Operations Research Management Science|Mathematics Interdisciplinary Applications|Social Sciences Mathematical Methods 3+4 Proceedings Paper Economics|Operations Research Management Science|Mathematics Interdisciplinary Applications|Social Sciences Mathematical Methods 2017 000427151400050 WOS cav_un_epca*0477966 Proceedings of the 35th International Conference Mathematical Methods in Economics (MME 2017) 978-80-7435-678-0 289 294 Hradec Králové University of Hradec Králové 2017