On Learning Mixture of Linear Regressions in the Non-Realizable Setting

Soumyabrata Pal, Arya Mazumdar, Rajat Sen, Avishek Ghosh
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:17202-17220, 2022.

Abstract

While mixture of linear regressions (MLR) is a well-studied topic, prior works usually do not analyze such models for prediction error. In fact, prediction and loss are not well-defined in the context of mixtures. In this paper, first we show that MLR can be used for prediction where instead of predicting a label, the model predicts a list of values (also known as list-decoding). The list size is equal to the number of components in the mixture, and the loss function is defined to be minimum among the losses resulted by all the component models. We show that with this definition, a solution of the empirical risk minimization (ERM) achieves small probability of prediction error. This begs for an algorithm to minimize the empirical risk for MLR, which is known to be computationally hard. Prior algorithmic works in MLR focus on the realizable setting, i.e., recovery of parameters when data is probabilistically generated by a mixed linear (noisy) model. In this paper we show that a version of the popular expectation minimization (EM) algorithm finds out the best fit lines in a dataset even when a realizable model is not assumed, under some regularity conditions on the dataset and the initial points, and thereby provides a solution for the ERM. We further provide an algorithm that runs in polynomial time in the number of datapoints, and recovers a good approximation of the best fit lines. The two algorithms are experimentally compared.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-pal22b, title = {On Learning Mixture of Linear Regressions in the Non-Realizable Setting}, author = {Pal, Soumyabrata and Mazumdar, Arya and Sen, Rajat and Ghosh, Avishek}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {17202--17220}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/pal22b/pal22b.pdf}, url = {https://proceedings.mlr.press/v162/pal22b.html}, abstract = {While mixture of linear regressions (MLR) is a well-studied topic, prior works usually do not analyze such models for prediction error. In fact, prediction and loss are not well-defined in the context of mixtures. In this paper, first we show that MLR can be used for prediction where instead of predicting a label, the model predicts a list of values (also known as list-decoding). The list size is equal to the number of components in the mixture, and the loss function is defined to be minimum among the losses resulted by all the component models. We show that with this definition, a solution of the empirical risk minimization (ERM) achieves small probability of prediction error. This begs for an algorithm to minimize the empirical risk for MLR, which is known to be computationally hard. Prior algorithmic works in MLR focus on the realizable setting, i.e., recovery of parameters when data is probabilistically generated by a mixed linear (noisy) model. In this paper we show that a version of the popular expectation minimization (EM) algorithm finds out the best fit lines in a dataset even when a realizable model is not assumed, under some regularity conditions on the dataset and the initial points, and thereby provides a solution for the ERM. We further provide an algorithm that runs in polynomial time in the number of datapoints, and recovers a good approximation of the best fit lines. The two algorithms are experimentally compared.} }
Endnote
%0 Conference Paper %T On Learning Mixture of Linear Regressions in the Non-Realizable Setting %A Soumyabrata Pal %A Arya Mazumdar %A Rajat Sen %A Avishek Ghosh %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-pal22b %I PMLR %P 17202--17220 %U https://proceedings.mlr.press/v162/pal22b.html %V 162 %X While mixture of linear regressions (MLR) is a well-studied topic, prior works usually do not analyze such models for prediction error. In fact, prediction and loss are not well-defined in the context of mixtures. In this paper, first we show that MLR can be used for prediction where instead of predicting a label, the model predicts a list of values (also known as list-decoding). The list size is equal to the number of components in the mixture, and the loss function is defined to be minimum among the losses resulted by all the component models. We show that with this definition, a solution of the empirical risk minimization (ERM) achieves small probability of prediction error. This begs for an algorithm to minimize the empirical risk for MLR, which is known to be computationally hard. Prior algorithmic works in MLR focus on the realizable setting, i.e., recovery of parameters when data is probabilistically generated by a mixed linear (noisy) model. In this paper we show that a version of the popular expectation minimization (EM) algorithm finds out the best fit lines in a dataset even when a realizable model is not assumed, under some regularity conditions on the dataset and the initial points, and thereby provides a solution for the ERM. We further provide an algorithm that runs in polynomial time in the number of datapoints, and recovers a good approximation of the best fit lines. The two algorithms are experimentally compared.
APA
Pal, S., Mazumdar, A., Sen, R. & Ghosh, A.. (2022). On Learning Mixture of Linear Regressions in the Non-Realizable Setting. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:17202-17220 Available from https://proceedings.mlr.press/v162/pal22b.html.

Related Material