Solving the Baby Intuitions Benchmark with a Hierarchically Bayesian Theory of Mind
| Autor: | Zhi-Xuan, Tan, Gothoskar, Nishad, Pollok, Falk, Gutfreund, Dan, Tenenbaum, Joshua B., Mansinghka, Vikash K. |
|---|---|
| Rok vydání: | 2022 |
| Předmět: | |
| Druh dokumentu: | Working Paper |
| Popis: | To facilitate the development of new models to bridge the gap between machine and human social intelligence, the recently proposed Baby Intuitions Benchmark (arXiv:2102.11938) provides a suite of tasks designed to evaluate commonsense reasoning about agents' goals and actions that even young infants exhibit. Here we present a principled Bayesian solution to this benchmark, based on a hierarchically Bayesian Theory of Mind (HBToM). By including hierarchical priors on agent goals and dispositions, inference over our HBToM model enables few-shot learning of the efficiency and preferences of an agent, which can then be used in commonsense plausibility judgements about subsequent agent behavior. This approach achieves near-perfect accuracy on most benchmark tasks, outperforming deep learning and imitation learning baselines while producing interpretable human-like inferences, demonstrating the advantages of structured Bayesian models of human social cognition. Comment: 6 pages, 2 figures. Presented at the Robotics: Science and Systems 2022 Workshop on Social Intelligence in Humans and Robots |
| Databáze: | arXiv |
| Externí odkaz: |
|