Reinforcement learning

Learning by doing rather than from a dataset. There is nothing collected in advance. The system acts, gets scored on the result, and adjusts to earn more over time.

Use it where the right move depends on the situation and the payoff arrives later: pricing, scheduling, routing, resource allocation, anything with a long chain of decisions. It needs somewhere safe to fail thousands of times first, which usually means a simulation, and that requirement is what keeps it rarer in business than the other two.