Fig. 4From: Inverse reinforcement learning for intelligent mechanical ventilation and sedative dosing in intensive care unitsThe convergence of P(O|R) using different initial values of the weightsBack to article page