PIPPS Flexible Model-Based Policy Search Robust to the Curse of Chaos

PIPPS:FlexibleModel-BasedPolicySearchRobusttotheCurseofChaosPaavoParmas1CarlEdwardRasmussen2JanPeters34KenjiDoya1AbstractVelocityPreviously,theexplodinggradientproblemhasPositionPositionbeenexplain...