fix over-parameterization of DQExperience

JuliaPOMDP / DeepQLearning.jl

Implementation of the Deep Q-learning algorithm to solve MDPs

Other

72 stars 13 forks source link

fix over-parameterization of DQExperience #64

Closed zsunberg closed 3 years ago

zsunberg commented 3 years ago

https://discourse.julialang.org/t/using-abstractenv-from-commonrlinterface-with-pomdps/68730/16

zsunberg commented 3 years ago

@MaximeBouton , I believe this fixes the bug in the link above. Ideally we would have a test that exercises this bug and the fix, but I don't have time to create it right now.

I have one question for you. This allows people to use Float64, which could cause performance degradation. My view is that having something that works is more important, but just wanted to see if you have a comment.