ResearchBenchmarksOffline Reinforcement Learning on half-me-kinematicOffline Reinforcement Learning on half-me-kinematicFollow46.3Performance ScoreDROCO37.14839.52441.944.276Dec 2, 2025Evaluation resultsMethodMethodLinksPerformance ScoreDROCOObservation Normalizat...Observation Normalization=false2025.1246.3IGDFObservation Normalizat...Observation Normalization=false2025.1240IQLObservation Normalizat...Observation Normalization=false2025.1238.9OTDFObservation Normalizat...Observation Normalization=false2025.1237.5