Learning about machine learning through tic-tac-toe competition scenarios
T.R. Crews, Ryan S. Boone · Journal of computing sciences in colleges · 2017
In the movie War Games, a global thermonuclear disaster was avoided because a computer program learned that tic-tac-toe has no winning move. But what assumptions about machine learning are necessary to get that result? This paper describes an experiment involving machine learning agents playing tic-tac-toe against each other using a variety of learning algorithms. The results indicate that the game tic-tac-toe has a distinct first mover advantage that influences the learning outcomes. The results also show that different learning strategies produce different learning results. These findings may be valuable to the discussion of machine learning in more complex environments.