Proceedings

On the Convergence of Stochastic Compositional Gradient Descent Ascent Method

Hongchang Gao, Xiaoqian Wang, Lei Luo, Xinghua Shi

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 2389-2395.

https://doi.org/10.24963/ijcai.2021/329

Keywords

  • Machine Learning: Cost-Sensitive Learning
  • Machine Learning: Adversarial Machine Learning

Abstract

The compositional minimax problem covers plenty of machine learning models such as the distributionally robust compositional optimization problem. However, it is yet another understudied problem to optimize the compositional minimax problem. In this paper, we develop a novel efficient stochastic compositional gradient descent ascent method for optimizing the compositional minimax problem. Moreover, we establish the theoretical convergence rate of our proposed method. To the best of our knowledge, this is the first work achieving such a convergence rate for the compositional minimax problem. Finally, we conduct extensive experiments to demonstrate the effectiveness of our proposed method.

Hongchang Gao, Xiaoqian Wang, Lei Luo, and Xinghua Shi. 2021. On the Convergence of Stochastic Compositional Gradient Descent Ascent Method. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 2389–2395. https://doi.org/10.24963/ijcai.2021/329
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards