Learning to Act by Predicting the Future
This addresses the problem of goal-free learning for AI agents in complex 3D simulations, though it is incremental as it builds on existing sensorimotor control paradigms.
The paper tackles sensorimotor control in immersive 3D environments by training a model to act based on raw sensory input without fixed goals, using a supervisory signal from cotemporal sensory and measurement streams. The approach outperforms prior methods in Doom simulations, winning the Visual Doom AI Competition in unseen environments.
We present an approach to sensorimotor control in immersive environments. Our approach utilizes a high-dimensional sensory stream and a lower-dimensional measurement stream. The cotemporal structure of these streams provides a rich supervisory signal, which enables training a sensorimotor control model by interacting with the environment. The model is trained using supervised learning techniques, but without extraneous supervision. It learns to act based on raw sensory input from a complex three-dimensional environment. The presented formulation enables learning without a fixed goal at training time, and pursuing dynamically changing goals at test time. We conduct extensive experiments in three-dimensional simulations based on the classical first-person game Doom. The results demonstrate that the presented approach outperforms sophisticated prior formulations, particularly on challenging tasks. The results also show that trained models successfully generalize across environments and goals. A model trained using the presented approach won the Full Deathmatch track of the Visual Doom AI Competition, which was held in previously unseen environments.