Abstract
The current researches for developing a Game Bot for digital games have been studied thoroughly. The bots created based on those studies play the game exactly as an expert would. The bots decide their next move by overlooking a number of moves ahead of the next move. Our approach of creating a Game Bot requires low computational resources without compromising the difficulty and expertise of play. We have proposed to develop our Game Bot through Reinforcement Learning. Reinforcement Learning is an exciting new field of machine learning, in which Bots learn by playing games. The bots are thrown in a gaming environment, and then trained to learn by observing their actions and rewards. We are going to build a Game Bot that will autonomously play against and beat the Atari game Neon Race Car.
Users
Please
log in to take part in the discussion (add own reviews or comments).