Towards a Principled Learning Rate Adaptation for Natural Evolution Strategies [slide]
This repository contains the code for the paper "Towards a Principled Learning Rate Adaptation for Natural Evolution Strategies". Please e-mail at [email protected] if you have any issue.
We use Poetry for our experimental environment. You can build the same environment by running the following commands:
# clone repository
$ git clone [email protected]:nomuramasahir0/xnes_adaptive_lr.git
# build the environment with poetry
$ cd xnes_adaptive_lr
$ poetry install
# run the experiment
$ poetry run python main.py sphereIf you find this code useful in your research then please cite:
@inproceedings{nomura2022towards,
title={Towards a Principled Learning Rate Adaptation for Natural Evolution Strategies},
author={Nomura, Masahiro and Ono, Isao},
booktitle={International Conference on the Applications of Evolutionary Computation (Part of EvoStar)},
pages={721--737},
year={2022},
organization={Springer}
}