Obstacle Tower: A Generalization Challenge in Vision, Control, and Planning
This provides a new generalization challenge for AI research, addressing the problem of robust performance in unseen environments for the field of reinforcement learning.
The authors introduced Obstacle Tower, a procedurally generated 3D benchmark requiring agents to solve low-level control and high-level planning from pixels with sparse rewards, and found that current deep RL methods and human players perform poorly on unseen instances.
The rapid pace of recent research in AI has been driven in part by the presence of fast and challenging simulation environments. These environments often take the form of games; with tasks ranging from simple board games, to competitive video games. We propose a new benchmark - Obstacle Tower: a high fidelity, 3D, 3rd person, procedurally generated environment. An agent playing Obstacle Tower must learn to solve both low-level control and high-level planning problems in tandem while learning from pixels and a sparse reward signal. Unlike other benchmarks such as the Arcade Learning Environment, evaluation of agent performance in Obstacle Tower is based on an agent's ability to perform well on unseen instances of the environment. In this paper we outline the environment and provide a set of baseline results produced by current state-of-the-art Deep RL methods as well as human players. These algorithms fail to produce agents capable of performing near human level.