EGD

Description : This package includes the Python code of the Evolutionary Gradient Descent (EGD) algorithm that combines typical components, i.e., population and mutation, of EAs with GD [1]. Experiments on synthetic functions and continuous MuJoCo locomotion tasks from the OpenAI Gym library [2], show the superior performance of EGD. README files are included in the package, showing how to use the code.

References: [1] Ke Xue, Chao Qian, Ling Xu, and Xudong Fei. Evolutionary Gradient Descent for Non-convex Optimization. In: Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI'21), Virtual, 2021, pp.3221-3227. [2] Greg Brockman, Vicki Cheung, Ludwig Pettersson, Jonas Schneider, John Schulman, Jie Tang, and Wojciech Zaremba. OpenAI Gym. CoRR abs/1606.01540, 2016.

ATTN: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact Dr. Chao Qian (qianc@lamda.nju.edu.cn).

Requirement: The package was developed with Python.

ATTN2: This package was developed by Mr. Ke Xue (xuek@lamda.nju.edu.cn). For any problem concerning the code, please feel free to contact Mr. Xue.

Download: code (1MB)