shivakanthsujit/reducible-loss
Codebase for Prioritizing samples in Reinforcement Learning with Reducible Loss
This project helps machine learning engineers and researchers improve how their reinforcement learning models learn from past experiences. It takes in historical training data from a reinforcement learning agent and prioritizes which samples the model should revisit, leading to more robust and efficient learning. The end user is a practitioner developing or deploying reinforcement learning systems.
No commits in the last 6 months.
Use this if you are training an off-policy Q-value reinforcement learning algorithm and want to optimize how your model prioritizes samples from its experience replay buffer, especially when dealing with noisy or stochastic data.
Not ideal if you are working with on-policy reinforcement learning algorithms or if your primary concern is not about optimizing sample prioritization in off-policy learning.
Stars
12
Forks
3
Language
Python
License
—
Category
Last pushed
Oct 10, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/shivakanthsujit/reducible-loss"
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