nnaisense/pgpelib
A mini library for Policy Gradients with Parameter-based Exploration, with reference implementation of the ClipUp optimizer (https://arxiv.org/abs/2008.02387) from NNAISENSE.
This library helps machine learning practitioners efficiently optimize complex systems using an approach inspired by how evolution works. It takes in a set of candidate solutions for a problem, evaluates how well each performs, and uses that feedback to intelligently propose improved solutions. This is ideal for those developing AI agents or tuning parameters in simulations.
No commits in the last 6 months.
Use this if you need to optimize parameters or develop policies for reinforcement learning problems, especially when derivative information is hard to get.
Not ideal if your optimization problem is simple and can be solved with traditional gradient-based methods.
Stars
73
Forks
4
Language
Python
License
BSD-3-Clause
Category
Last pushed
Dec 10, 2020
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/nnaisense/pgpelib"
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