CleanRL (@cleanrl_lib) 's Twitter Profile
CleanRL

@cleanrl_lib

High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

ID: 1531787789817331712

linkhttp://cleanrl.dev calendar_today01-06-2022 00:01:07

3 Tweet

637 Followers

0 Following

Costa Huang (@vwxyzjn) 's Twitter Profile Photo
Costa Huang
@vwxyzjn

3 years ago

Thanks to @_joaogui1's awesome contribution 🙏, CleanRL now has a TD3 + JAX implementation that is 2-4x faster than the TD3 + PyTorch equivalent 🔥. Running on TPU is now possible, too 🚀! 📜 docs: docs.cleanrl.dev/rl-algorithms/… 💾 code: github.com/vwxyzjn/cleanr… A short 🧵1/x

Thanks to @_joaogui1's awesome contribution 🙏, <a href="/cleanrl_lib/">CleanRL</a> now has a TD3 + JAX implementation that is 2-4x faster than the TD3 + <a href="/PyTorch/">PyTorch</a> equivalent 🔥. Running on TPU is now possible, too 🚀! 📜 docs: docs.cleanrl.dev/rl-algorithms/… 💾 code: github.com/vwxyzjn/cleanr… A short 🧵1/x
Chang Ye (@yooceii) 's Twitter Profile Photo
Chang Ye
@yooceii

3 years ago

Happy to share that CleanRL now supports Random Network Distillation + envpool, it's 3× faster than our first version without envpool and still have comparable performance to the original implementation, say 👋 to the long training time on hard-exploration games! Details👇

Happy to share that <a href="/cleanrl_lib/">CleanRL</a> now supports Random Network Distillation + envpool, it's 3× faster than our first version without envpool and still have comparable performance to the original implementation, say 👋 to the long training time on hard-exploration games! Details👇