Yarden As

la-mbda. LAMBDA is a model-based reinforcement learning agent that uses Bayesian world models for safe policy optimization

jax-dreamer. Dreamer on JAX

actsafe. Scaling safe exploration to vision control

panda-rl-kit. Deploy RL on your Real-World Franka Emika Panda

probabilistic-regression. A short tutorial on Bayesian Neural Networks with JAX

ethz-safe-learning. Safe Model-Based Reinforcement Learning via Ensembles and Model Predictive Control

yas.pub. Astro

jax-pacoh-nn. Jupyter Notebook

talks. My Slidev slides

safe-ef. Code for “Safe-EF: Error Feedback for Nonsmooth Constrained Optimization”