Decision-Focused Learning: Through the Lens of Learning to Rank

Jayanta Mandi, Vı́ctor Bucarey, Maxime Mulamba Ke Tchomba, Tias Guns
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:14935-14947, 2022.

Abstract

In the last years decision-focused learning framework, also known as predict-and-optimize, have received increasing attention. In this setting, the predictions of a machine learning model are used as estimated cost coefficients in the objective function of a discrete combinatorial optimization problem for decision making. Decision-focused learning proposes to train the ML models, often neural network models, by directly optimizing the quality of decisions made by the optimization solvers. Based on a recent work that proposed a noise contrastive estimation loss over a subset of the solution space, we observe that decision-focused learning can more generally be seen as a learning-to-rank problem, where the goal is to learn an objective function that ranks the feasible points correctly. This observation is independent of the optimization method used and of the form of the objective function. We develop pointwise, pairwise and listwise ranking loss functions, which can be differentiated in closed form given a subset of solutions. We empirically investigate the quality of our generic methods compared to existing decision-focused learning approaches with competitive results. Furthermore, controlling the subset of solutions allows controlling the runtime considerably, with limited effect on regret.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-mandi22a, title = {Decision-Focused Learning: Through the Lens of Learning to Rank}, author = {Mandi, Jayanta and Bucarey, V\'{\i}ctor and Tchomba, Maxime Mulamba Ke and Guns, Tias}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {14935--14947}, 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/mandi22a/mandi22a.pdf}, url = {https://proceedings.mlr.press/v162/mandi22a.html}, abstract = {In the last years decision-focused learning framework, also known as predict-and-optimize, have received increasing attention. In this setting, the predictions of a machine learning model are used as estimated cost coefficients in the objective function of a discrete combinatorial optimization problem for decision making. Decision-focused learning proposes to train the ML models, often neural network models, by directly optimizing the quality of decisions made by the optimization solvers. Based on a recent work that proposed a noise contrastive estimation loss over a subset of the solution space, we observe that decision-focused learning can more generally be seen as a learning-to-rank problem, where the goal is to learn an objective function that ranks the feasible points correctly. This observation is independent of the optimization method used and of the form of the objective function. We develop pointwise, pairwise and listwise ranking loss functions, which can be differentiated in closed form given a subset of solutions. We empirically investigate the quality of our generic methods compared to existing decision-focused learning approaches with competitive results. Furthermore, controlling the subset of solutions allows controlling the runtime considerably, with limited effect on regret.} }
Endnote
%0 Conference Paper %T Decision-Focused Learning: Through the Lens of Learning to Rank %A Jayanta Mandi %A Vı́ctor Bucarey %A Maxime Mulamba Ke Tchomba %A Tias Guns %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-mandi22a %I PMLR %P 14935--14947 %U https://proceedings.mlr.press/v162/mandi22a.html %V 162 %X In the last years decision-focused learning framework, also known as predict-and-optimize, have received increasing attention. In this setting, the predictions of a machine learning model are used as estimated cost coefficients in the objective function of a discrete combinatorial optimization problem for decision making. Decision-focused learning proposes to train the ML models, often neural network models, by directly optimizing the quality of decisions made by the optimization solvers. Based on a recent work that proposed a noise contrastive estimation loss over a subset of the solution space, we observe that decision-focused learning can more generally be seen as a learning-to-rank problem, where the goal is to learn an objective function that ranks the feasible points correctly. This observation is independent of the optimization method used and of the form of the objective function. We develop pointwise, pairwise and listwise ranking loss functions, which can be differentiated in closed form given a subset of solutions. We empirically investigate the quality of our generic methods compared to existing decision-focused learning approaches with competitive results. Furthermore, controlling the subset of solutions allows controlling the runtime considerably, with limited effect on regret.
APA
Mandi, J., Bucarey, V., Tchomba, M.M.K. & Guns, T.. (2022). Decision-Focused Learning: Through the Lens of Learning to Rank. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:14935-14947 Available from https://proceedings.mlr.press/v162/mandi22a.html.

Related Material