AurelianTactics

Thanks for the feedback. I'll try to put something together. > if I get it, the proposed algorithm just changes the loss function but nothing else? In that case, I...

I'm still working on this. I had a few questions. I was able to get the implementation of DQNReg and DQNClipped to work in a contrib fork (https://github.com/AurelianTactics/stable-baselines3-contrib) on the...

You don't want the lib package that you install from pip. You want the lib directory that comes in dennybritz repo. You can find that lib folder here: https://github.com/dennybritz/reinforcement-learning/tree/master/lib