Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations

ExtrapolatingBeyondSuboptimalDemonstrationsviaInverseReinforcementLearningfromObservationsDanielS.Brown1WonjoonGoo1PrabhatNagarajan2ScottNiekum1AbstractFigure1.T-REXtakesasequenceofrankeddemonstrat...