Happy to share that @cleanrl_lib now supports Random Network Distillation + envpool, it's 3ร faster than our first version without envpool and still have comparable performance to the original implementation, say ๐ to the long training time on hard-exploration games!
Details๐
Thanks to @_joaogui1's awesome contribution ๐, @cleanrl_lib now has a TD3 + JAX implementation that is 2-4x faster than the TD3 + @PyTorch equivalent ๐ฅ. Running on TPU is now possible, too ๐!
๐ docs: https://t.co/JXXh3xMGjY
๐พ code: https://t.co/VXBnq4sh8G
A short ๐งต1/x