EWC for reinforcement learning

[g277321]

Hi,
i have a question for your code. How does this DQN with EWC work for reinforcement learning?

[ghost]

Hello. The DQN parts were not done by me and were added by @irakli97 as an experiment, They're not "officially" a part of our internal framework. I'll hit @irakli97 up and he might be able to help.

[ghost]

Irakli got back to me. He never finished the DQN part. We might do that in a few days and add it back. In the meantime, I've removed the experimental code.