Transformer Based Reinforcement Learning For Games
This work addresses the challenge of game-playing AI, but it appears incremental as it adapts a known NLP method (transformer) to a new domain without claiming major breakthroughs.
The paper tackles the problem of training agents to play complex video games by proposing a transformer-based method, comparing it with existing deep reinforcement learning techniques that use recurrent neural networks.
Recent times have witnessed sharp improvements in reinforcement learning tasks using deep reinforcement learning techniques like Deep Q Networks, Policy Gradients, Actor Critic methods which are based on deep learning based models and back-propagation of gradients to train such models. An active area of research in reinforcement learning is about training agents to play complex video games, which so far has been something accomplished only by human intelligence. Some state of the art performances in video game playing using deep reinforcement learning are obtained by processing the sequence of frames from video games, passing them through a convolutional network to obtain features and then using recurrent neural networks to figure out the action leading to optimal rewards. The recurrent neural network will learn to extract the meaningful signal out of the sequence of such features. In this work, we propose a method utilizing a transformer network which have recently replaced RNNs in Natural Language Processing (NLP), and perform experiments to compare with existing methods.