Skip to main content
fantomas
Projects
rl-with-il-reward
Reports
Negatives from Experience Replay vs Negatives from Expert

Negatives from Experience Replay vs Negatives from Expert

Size of Experience Replay is 10. Fixed subset of size 10 is chosen from the expert dataset.
Artem Tsypin
Created on March 12|Last edited on March 12
Environment reward is plotted as AverageEnvEpRet.

Computing group metrics from first 10 groups
0100200300400Step-200-180-160-140-120-100
Computing group metrics from first 10 groups
0100200300400Step50100150
Computing group metrics from first 10 groups
0100200300400Step-250-200-150-100-500
Computing group metrics from first 10 groups
0100200300400Step-800-600-400-200
Computing group metrics from first 10 groups
0100200300400Step-200-1000100
Computing group metrics from first 10 groups
0100200300400Step-800-600-400-200
Run set
329


Created with ❤️ on Weights & Biases.

https://wandb.ai/fantomas/rl-with-il-reward/reports/Negatives-from-Experience-Replay-vs-Negatives-from-Expert---Vmlldzo1MjQ0NTA
Made with Weights & Biases. Sign up or log in to create reports like this one.