Fig. 5From: Inverse reinforcement learning for intelligent mechanical ventilation and sedative dosing in intensive care unitsComparison of feature importance using different RL and IRL policiesBack to article page