Reinforcement Learning · Interactive activity

MDPs, Bellman and value iteration

0 XP

₹100 now or ₹150 next year?

Most decisions pay off later. A chess move can look bad now and win the game 20 moves on.

How should a machine value the future?