ICML · 2024

Building Socially-Equitable Public Models

Yejia Liu, Jianyi Yang, Pengfei Li, Tongxin Li, Shaolei Ren

Proceedings of the International Conference on Machine Learning

Read paper Repository BibTeX

Who benefits when many decision makers share one predictive model?

The Equitable Objective trains a public predictive model around the outcomes of heterogeneous downstream agents. A policy gradient method and theoretical analysis study a more equitable distribution of performance, beyond optimizing prediction accuracy alone.

Research themes

Connections

Coordination through information and incentives

Virtual power plants use renewable predictions to coordinate consumers. Energy sharing and transportation games model how charging demand, prices, and network constraints interact.

Open questions

Accountable learning across agents

How should shared models balance efficiency, strategic risk, and equity?

Explore open directions

Cite this paper

Yejia Liu, Jianyi Yang, Pengfei Li, Tongxin Li, Shaolei Ren. Building Socially-Equitable Public Models. Proceedings of the International Conference on Machine Learning, 2024. https://proceedings.mlr.press/v235/liu24bw.html

BibTeX
@inproceedings{tongxin-equitable-models,
  title = {{Building Socially-Equitable Public Models}},
  author = {Yejia Liu and Jianyi Yang and Pengfei Li and Tongxin Li and Shaolei Ren},
  year = {2024},
  booktitle = {Proceedings of the International Conference on Machine Learning},
  url = {https://proceedings.mlr.press/v235/liu24bw.html}
}