Posts

Comments

Comment by araffin on Jitters No Evidence of Stupidity in RL · 2021-09-19T19:00:31.234Z · LW · GW

You may have a look at "Smooth Exploration for Robotic Reinforcement Learning" ;) The jitter issue is one of the main motivation of that paper: https://openreview.net/forum?id=TSuSGVkjuXd

But overall, energy minimization is a good regulariser.

Also related: https://openreview.net/forum?id=PfC1Jr6gvuP