evgenii-nikishin/omd
JAX code for the paper "Control-Oriented Model-Based Reinforcement Learning with Implicit Differentiation"
This project offers a method for training machine learning models that control physical systems or agents. It takes in raw data from interactions with an environment, like a robot learning to walk or a system learning to balance, and produces a more effective control policy. Researchers and engineers working on reinforcement learning problems would use this to improve the performance of their learned agents.
No commits in the last 6 months.
Use this if you are a researcher or practitioner in reinforcement learning looking to develop models that directly optimize for cumulative rewards rather than just predicting future states.
Not ideal if you are looking for a general-purpose machine learning library or a tool for supervised learning tasks unrelated to control problems.
Stars
44
Forks
6
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jun 14, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/evgenii-nikishin/omd"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
explosion/thinc
🔮 A refreshing functional take on deep learning, compatible with your favorite libraries
google-deepmind/optax
Optax is a gradient processing and optimization library for JAX.
patrick-kidger/diffrax
Numerical differential equation solvers in JAX. Autodifferentiable and GPU-capable....
google/grain
Library for reading and processing ML training data.
patrick-kidger/equinox
Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/