How to implement the baseline EIPO in this Project?

[Charlie0257]

Hi, @williamd4112

Thanks for your great work HEPO!
I want to know the command for implementing the baseline EIPO because I noticed self.use_switch in the project:)

Thanks for any advice!
Best,
Charlie

[williamd4112]

Hi Charlie,

We will add EIPO to the current codebase soon. Please let me know if you have other questions and feel free to remind me at zwhong@mit.edu

[Charlie0257]

@williamd4112 Thanks for your reply!

When I trained my task with HEPO, I noticed a performance collapse in HEPO's performance, not as good as ref policy, much less heuristics.

I don't know if you have encountered this when tuning your hyperparameters.

Can you offer any advice on hyperparameter tuning or anything else?

Best,
Charlie