Path Integral Policy Improvement With Population Adaptation.
Sign Up to like & getrecommendations! Published in 2020 at "IEEE transactions on cybernetics"
DOI: 10.1109/tcyb.2020.2983923
Abstract: Path integral policy improvement (PI²) is known to be an efficient reinforcement learning algorithm, particularly, if the target system is a high-dimensional dynamical system. However, PI², and its existing extensions, have adjustable parameters, on which… read more here.
Keywords: integral policy; path integral; policy improvement; improvement population ... See more keywords