search

kyonofx / tia

[ICML 2021] Learning Task Informed Abstractions -- a representation learning approach for model-based RL in complex visual domains
https://xiangfu.co/tia
17 stars 3 forks source link

Evaulation graphs for DMC #3

Closed vedantdave97 closed 1 year ago

vedantdave97 commented 1 year ago

Hello @kyonofx, Thanks for the paper and the training/evaluation graphs.

In the google drive link provided, I can only find graphs for Dreamer in DMC environment. Can you please provide the graphs for TIA in the standard environment (no noise)? (I am talking about Figure 10, the blue plot with "no Bg")

If I am missing something, let me know.

kyonofx commented 1 year ago

Hi,

The google drive folder contains metrics for TIA on the driving video tasks. We did not run experiments of TIA on clean environments (TIA will not help in clean environments). Figure 10 reports the performance of Dreamer, for which all metrics are in the google drive.

vedantdave97 commented 1 year ago

Hello,

Thanks for the quick response and clarification.