notesum.ai

Published at October 30

cs.LG

cs.AI

cs.RO

stat.ML

Released Date: October 30, 2024

Authors: Ruhan Wang¹, Yu Yang¹, Zhishuai Liu¹, Dongruo Zhou¹, Pan Xu¹

Aff.: ¹Institution Not Specified

Shift

Environment

BEAR

D-BEAR

AWR

D-AWR

BCQ

D-BCQ

CQL

D-CQL

MOPO

D-MOPO

Walker2D

\textbf{5.776}_{\pm\textbf{1.653}}

\textbf{6.516}_{\pm\textbf{3.220}}

\textbf{77.442}_{\pm\textbf{0.340}}

\textbf{78.004}_{\pm\textbf{0.911}}

\textbf{70.681}_{\pm\textbf{0.539}}

\textbf{72.023}_{\pm\textbf{0.695}}

\textbf{73.317}_{\pm\textbf{1.368}}

\textbf{74.276}_{\pm\textbf{2.582}}

\textbf{21.617}_{\pm\textbf{1.277}}

\textbf{21.621}_{\pm\textbf{1.063}}

\textbf{78.768}_{\pm\textbf{1.233}}

Hopper

\textbf{22.436}_{\pm\textbf{0.103}}

\textbf{25.608}_{\pm\textbf{1.063}}

\textbf{25.843}_{\pm\textbf{0.325}}

\textbf{26.594}_{\pm\textbf{1.267}}

\textbf{24.853}_{\pm\textbf{1.615}}

\textbf{26.487}_{\pm\textbf{1.366}}

\textbf{49.094}_{\pm\textbf{2.207}}

\textbf{45.101}_{\pm\textbf{0.342}}

\textbf{20.765}_{\pm\textbf{3.350}}

\textbf{21.495}_{\pm\textbf{0.848}}

\textbf{34.057}_{\pm\textbf{0.177}}

HalfCheetah

\textbf{5.431}_{\pm\textbf{1.518}}

\textbf{6.009}_{\pm\textbf{1.705}}

\textbf{42.293}_{\pm\textbf{0.862}}

\textbf{41.800}_{\pm\textbf{0.830}}

\textbf{39.835}_{\pm\textbf{0.427}}

\textbf{39.333}_{\pm\textbf{0.506}}

\textbf{37.081}_{\pm\textbf{0.358}}

\textbf{37.189}_{\pm\textbf{0.218}}

\textbf{58.457}_{\pm\textbf{1.449}}

\textbf{59.311}_{\pm\textbf{0.949}}

\textbf{46.954}_{\pm\textbf{0.260}}

Walker2D

\textbf{4.926}_{\pm\textbf{1.418}}

\textbf{6.933}_{\pm\textbf{1.884}}

\textbf{67.636}_{\pm\textbf{1.468}}

\textbf{64.303}_{\pm\textbf{0.513}}

\textbf{62.696}_{\pm\textbf{1.307}}

\textbf{60.681}_{\pm\textbf{1.118}}

\textbf{68.962}_{\pm\textbf{0.865}}

\textbf{69.141}_{\pm\textbf{0.944}}

\textbf{23.552}_{\pm\textbf{1.063}}

\textbf{23.574}_{\pm\textbf{0.665}}

{\color[rgb]{0,0,1}\definecolor[named]{pgfstrokecolor}{rgb}{0,0,1}\textbf{71.0% 68}_{\pm\textbf{1.022}}}

Hopper

\textbf{8.536}_{\pm\textbf{1.965}}

\textbf{10.576}_{\pm\textbf{2.052}}

\textbf{57.021}_{\pm\textbf{0.938}}

\textbf{61.463}_{\pm\textbf{0.702}}

\textbf{74.559}_{\pm\textbf{0.605}}

\textbf{74.853}_{\pm\textbf{0.626}}

\textbf{71.495}_{\pm\textbf{0.126}}

\textbf{63.611}_{\pm\textbf{1.136}}

\textbf{23.556}_{\pm\textbf{1.327}}

\textbf{24.992}_{\pm\textbf{0.944}}

\textbf{70.685}_{\pm\textbf{0.726}}

HalfCheetah

\textbf{1.948}_{\pm\textbf{1.058}}

\textbf{2.901}_{\pm\textbf{0.402}}

\textbf{41.992}_{\pm\textbf{0.762}}

\textbf{42.545}_{\pm\textbf{0.731}}

\textbf{50.511}_{\pm\textbf{0.371}}

\textbf{52.149}_{\pm\textbf{0.457}}

\textbf{49.046}_{\pm\textbf{0.420}}

\textbf{49.284}_{\pm\textbf{0.570}}

\textbf{61.073}_{\pm\textbf{0.315}}

{\color[rgb]{0,0,1}\definecolor[named]{pgfstrokecolor}{rgb}{0,0,1}\textbf{61.4% 47}_{\pm\textbf{0.734}}}

\textbf{47.725}_{\pm\textbf{0.431}}