Towards Balanced Representation Learning for Credit Policy Evaluation

Yiyan Huang, Cheuk Hang Leung, Shumin Ma, Zhiri Yuan, Qi Wu, Siyi Wang, Dongdong Wang, Zhixiang Huang
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:3677-3692, 2023.

Abstract

Credit policy evaluation presents profitable opportunities for E-commerce platforms through improved decision-making. The core of policy evaluation is estimating the causal effects of the policy on the target outcome. However, selection bias presents a key challenge in estimating causal effects from real-world data. Some recent causal inference methods attempt to mitigate selection bias by leveraging covariate balancing in the representation space to obtain the domain-invariant features. However, it is noticeable that balanced representation learning can be accompanied by a failure of domain discrimination, resulting in the loss of domain-related information. This is referred to as the over-balancing issue. In this paper, we introduce a novel objective for representation balancing methods to do policy evaluation. In particular, we construct a doubly robust loss based on the predictions of treatment and outcomes, serving as a prerequisite for covariate balancing to deal with the over-balancing issue. In addition, we investigate how to improve treatment effect estimations by exploiting the unconfoundedness assumption. The extensive experimental results on benchmark datasets and a newly introduced credit dataset show a general outperformance of our method compared with existing methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v206-huang23b, title = {Towards Balanced Representation Learning for Credit Policy Evaluation}, author = {Huang, Yiyan and Leung, Cheuk Hang and Ma, Shumin and Yuan, Zhiri and Wu, Qi and Wang, Siyi and Wang, Dongdong and Huang, Zhixiang}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {3677--3692}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/huang23b/huang23b.pdf}, url = {https://proceedings.mlr.press/v206/huang23b.html}, abstract = {Credit policy evaluation presents profitable opportunities for E-commerce platforms through improved decision-making. The core of policy evaluation is estimating the causal effects of the policy on the target outcome. However, selection bias presents a key challenge in estimating causal effects from real-world data. Some recent causal inference methods attempt to mitigate selection bias by leveraging covariate balancing in the representation space to obtain the domain-invariant features. However, it is noticeable that balanced representation learning can be accompanied by a failure of domain discrimination, resulting in the loss of domain-related information. This is referred to as the over-balancing issue. In this paper, we introduce a novel objective for representation balancing methods to do policy evaluation. In particular, we construct a doubly robust loss based on the predictions of treatment and outcomes, serving as a prerequisite for covariate balancing to deal with the over-balancing issue. In addition, we investigate how to improve treatment effect estimations by exploiting the unconfoundedness assumption. The extensive experimental results on benchmark datasets and a newly introduced credit dataset show a general outperformance of our method compared with existing methods.} }
Endnote
%0 Conference Paper %T Towards Balanced Representation Learning for Credit Policy Evaluation %A Yiyan Huang %A Cheuk Hang Leung %A Shumin Ma %A Zhiri Yuan %A Qi Wu %A Siyi Wang %A Dongdong Wang %A Zhixiang Huang %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-huang23b %I PMLR %P 3677--3692 %U https://proceedings.mlr.press/v206/huang23b.html %V 206 %X Credit policy evaluation presents profitable opportunities for E-commerce platforms through improved decision-making. The core of policy evaluation is estimating the causal effects of the policy on the target outcome. However, selection bias presents a key challenge in estimating causal effects from real-world data. Some recent causal inference methods attempt to mitigate selection bias by leveraging covariate balancing in the representation space to obtain the domain-invariant features. However, it is noticeable that balanced representation learning can be accompanied by a failure of domain discrimination, resulting in the loss of domain-related information. This is referred to as the over-balancing issue. In this paper, we introduce a novel objective for representation balancing methods to do policy evaluation. In particular, we construct a doubly robust loss based on the predictions of treatment and outcomes, serving as a prerequisite for covariate balancing to deal with the over-balancing issue. In addition, we investigate how to improve treatment effect estimations by exploiting the unconfoundedness assumption. The extensive experimental results on benchmark datasets and a newly introduced credit dataset show a general outperformance of our method compared with existing methods.
APA
Huang, Y., Leung, C.H., Ma, S., Yuan, Z., Wu, Q., Wang, S., Wang, D. & Huang, Z.. (2023). Towards Balanced Representation Learning for Credit Policy Evaluation. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:3677-3692 Available from https://proceedings.mlr.press/v206/huang23b.html.

Related Material