Map Induction: Compositional spatial submap learning for efficient exploration in novel environments
This work addresses the challenge of improving exploration algorithms for AI systems by drawing insights from human cognition, though it is incremental in applying existing program induction methods to a new domain.
The paper tackled the problem of efficient exploration in novel environments by modeling human cognitive mechanisms for inferring unobserved spatial structure, and demonstrated that their computational framework outperforms state-of-the-art planning algorithms in a realistic navigation domain.
Humans are expert explorers. Understanding the computational cognitive mechanisms that support this efficiency can advance the study of the human mind and enable more efficient exploration algorithms. We hypothesize that humans explore new environments efficiently by inferring the structure of unobserved spaces using spatial information collected from previously explored spaces. This cognitive process can be modeled computationally using program induction in a Hierarchical Bayesian framework that explicitly reasons about uncertainty with strong spatial priors. Using a new behavioral Map Induction Task, we demonstrate that this computational framework explains human exploration behavior better than non-inductive models and outperforms state-of-the-art planning algorithms when applied to a realistic spatial navigation domain.