R2D2
R2D2 copied to clipboard
Published
20 hours ago •
ZiyuanMa
ZiyuanMa
An Implementation of Recurrent Experience Replay in Distributed Reinforcement Learning (Kapturowski et al. 2019) in PyTorch
Results
2
R2D2 issues
Sort by
recently updated
recently updated
newest added
I tried this r2d2 with Gravitar instead of Pacman, and it didn't seem like it was learning. I even increased the buffer limit to 1_000_000 .
As a beginner in reinforcement learning, how should I debug this multi-process and multi-thread project? In this way, I can understand R2D2 better and make certain modifications. Thanks for your...
