See, Hear, Explore: Curiosity via Audio-Visual Association
This work addresses the problem of exploration for reinforcement learning agents, offering a novel method that leverages multi-modal associations to improve efficiency, though it is incremental in building on existing curiosity formulations.
The paper tackles the challenge of exploration in reinforcement learning by introducing a curiosity-driven approach that rewards novel associations between audio and visual senses, leading to more efficient exploration in Atari environments and Habitat without external rewards.
Exploration is one of the core challenges in reinforcement learning. A common formulation of curiosity-driven exploration uses the difference between the real future and the future predicted by a learned model. However, predicting the future is an inherently difficult task which can be ill-posed in the face of stochasticity. In this paper, we introduce an alternative form of curiosity that rewards novel associations between different senses. Our approach exploits multiple modalities to provide a stronger signal for more efficient exploration. Our method is inspired by the fact that, for humans, both sight and sound play a critical role in exploration. We present results on several Atari environments and Habitat (a photorealistic navigation simulator), showing the benefits of using an audio-visual association model for intrinsically guiding learning agents in the absence of external rewards. For videos and code, see https://vdean.github.io/audio-curiosity.html.