|
This article is cited in 3 scientific papers (total in 3 papers)
Optimization, System Analysis, and Operations Research
Probabilistic interpretation of the distillation problem
A. V. Grabovoya, V. V. Strijovab a Moscow Institute of Physics and Technology, Dolgoprudnyi, Moscow oblast, 141701 Russia
b Dorodnicyn Computing Centre, Russian Academy of Sciences, Moscow, 119333 Russia
Abstract:
The article deals with methods for reducing the complexity of approximating models. Probabilistic substantiation of distillation and privileged teaching methods is proposed. General conclusions are given for an arbitrary parametric function with a predetermined structure. A theoretical basis is demonstrated for the special cases of linear and logistic regression. The analysis of the considered models is carried out in a computational experiment on synthetic samples and real data. The FashionMNIST and Twitter Sentiment Analysis samples are considered as real data.
Keywords:
model selection, Bayesian inference, model distillation, learning with privileged information.
Citation:
A. V. Grabovoy, V. V. Strijov, “Probabilistic interpretation of the distillation problem”, Avtomat. i Telemekh., 2022, no. 1, 150–168; Autom. Remote Control, 83:1 (2022), 123–137
Linking options:
https://www.mathnet.ru/eng/at15892 https://www.mathnet.ru/eng/at/y2022/i1/p150
|
Statistics & downloads: |
Abstract page: | 127 | Full-text PDF : | 1 | References: | 13 | First page: | 16 |
|