a modular reinforcement learning library with JAX agents
Multi-task reinforcement learning framework for agents to perform goal-conditioned tasks using end-effector control with Franka Emika Arm
Jax-Based Off-Policy RL Algorithms
Sim-to-real RL for all 56 equilibrium transitions of a triple inverted pendulum on a cart. MuJoCo + TQC + n8n-orchestrated pipeline.
BipedalWalker: Classic 最速 & Hardcore も速く（5人チーム / TQC / Kaggle + W&B）
A 20-DOF humanoid robot learning to stand from lying positions using TQC (Truncated Quantile Critics) reinforcement learning. Features custom JAX implementation with 5 quantile critics, asymmetric actor-critic observations, and contact-rich simulation.
Reinforcement learning project using Truncated Quantile Critics (TQC) with elite experience replay and observation normalisation to train a Bipedal Walker agent. Includes report, training code, logs, and videos of best runs for both standard and hardcore environments.
Add a description, image, and links to the tqc topic page so that developers can more easily learn about it.
To associate your repository with the tqc topic, visit your repo's landing page and select "manage topics."