Skip to main content
openrlbenchmark
Projects
sbx
Reports
PPO CNN - Performance report

PPO CNN - Performance report

Antonin RAFFIN
Created on September 29|Last edited on September 29
Note: SBX PPO is using a copy hack to share the features extractor between the actor and the critic (without it, performances are much worse when the CNN is not shared)

PongNoFrameskip-v4



2M4M6M8M10MTimesteps-20-1001020Mean Episodic Return
algo: ppo SBX
algo: ppo SB3
SBX
2
SB3
6


BreakoutNoFrameskip-v4



SBX
3
SB3
6


Created with ❤️ on Weights & Biases.

https://wandb.ai/openrlbenchmark/sbx/reports/PPO-CNN-Performance-report--VmlldzoxNDU2OTk2OA
Made with Weights & Biases. Sign up or log in to create reports like this one.