Lessons from Developing an AI to Play Magic: The Gathering by Melvin Zhang

Published on: Saturday, 3 October 2015

Although computers have beaten humans at Chess, Checkers, and even Jeopardy, games with complex board states (such as Go) or hidden information (such as Poker) have remained out of reach. Magic: the Gathering is a strategy card game that has both complex board states and hidden information.

Magarena's goal is to develop an open source competitive computer MTG player. This talk distills some of the lessons learnt in developing an AI that is able to make decisions under uncertainly.

This talk was recorded at GeekCampSG 2015 on 26 September 2015.

Visit http://geekcamp.sg for more information about the event