goktug97/PEPG-ES
Python Implementation of Parameter-exploring Policy Gradients Evolution Strategy
This project helps machine learning practitioners optimize the parameters of neural networks using an evolution strategy called Parameter-exploring Policy Gradients (PEPG). You input a neural network's architecture and a reward function, and it outputs the best set of network parameters to achieve a high reward. This is primarily useful for those working on reinforcement learning or black-box optimization problems.
No commits in the last 6 months. Available on PyPI.
Use this if you need to optimize parameters for a neural network, especially in reinforcement learning contexts, and want an alternative to gradient-based methods like backpropagation.
Not ideal if you are working on supervised learning problems where backpropagation is generally faster and more reliable for training neural networks.
Stars
17
Forks
—
Language
Python
License
MIT
Category
Last pushed
Apr 02, 2020
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/goktug97/PEPG-ES"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
DLR-RM/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
google-deepmind/dm_control
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning...
Denys88/rl_games
RL implementations
pytorch/rl
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
yandexdataschool/Practical_RL
A course in reinforcement learning in the wild