Reinforcementlearning.jjl

[ttidrick]

Hi All,
I’m curious if anyone has opinions, comments, heads-ups, etc. concerning the package ‘ReinforcementLearning.jl’.
Thanks,
Ty Tidrick

[baggepinnen]

The issue list in the repository is a kind of “heads-up”

[WuSiren]

I tried this package more than one year ago and felt its then documentation was not as friendly as at least I a newbie wish. I’ve asked a question here at that time and still been waiting for the first response so far, though I’ve almost forgotten what I was doing then. Maybe the question is too trivial?

How to perform an Off-Policy reinforcement learning using `ReinforcementLearning.jl`?

I am a beginner of both Julia and RL. Now I’m trying to training a Q-based tabular policy with off-policy strategy using ReinforcementLearning.jl. Let behavior_policy be a RandomPolicy (for example) to collect experiences, I want the target_policy as a QBasedPolicy to be trained. I wrote the following code: target_policy = QBasedPolicy( learner = MonteCarloLearner(; approximator=TabularQApproximator( ;n_state = length(state_space(wrapped_env)), n_action = len…

[colintbowers]

I’m not an expert by any means, but I believe I read somewhere that they are currently doing a fairly extensive re-write of a lot of the package, and that consequently, if you find any bugs or issues in the existing package, you’re likely stuck with them until the current refactoring is done.