Learning to Play the Game of Nim with Reinforcement Learning (Q-learning)

Описание: Learning Project 4b from CS50 AI (by Harvard @edX): Learning to play the Game of Nim with Reinforcement Learning (Q-learning), the AI agent plays 10000 games with itself and learns from the rewards / its mistakes, by updating the q-values with the equation

Q(s,a)=Q(s,a) + α.(R(s,a) + γ.(max {a′} Q(s′,a′ )) − Q(s,a))

in the initial rounds of training.

As can be seen, the q-values are learnt for only a few (state, action) pairs, since almost all of the state-spaces are yet to be explored by the agent.

(#python #ml #reinforcementlearning #ai)