Sequential experiments

optimization technique for undifferentiable, black-box functions

reinforcement learning problem exemplifying the exploration–exploitation tradeoff

sequential experimentation for improvement and for finding an optimal response