TL;DR
Causal-PIK introduces a physics-informed, causality-based approach using Bayesian optimization to improve physical reasoning and planning efficiency in complex object interaction tasks, outperforming state-of-the-art methods.
Contribution
It presents a novel causality-based method with a physics-informed kernel that enhances planning efficiency in physical reasoning tasks.
Findings
Outperforms state-of-the-art on Virtual Tools and PHYRE benchmarks.
Requires fewer actions to reach goals compared to previous methods.
Remains competitive with human problem-solvers on challenging tasks.
Abstract
Tasks that involve complex interactions between objects with unknown dynamics make planning before execution difficult. These tasks require agents to iteratively improve their actions after actively exploring causes and effects in the environment. For these type of tasks, we propose Causal-PIK, a method that leverages Bayesian optimization to reason about causal interactions via a Physics-Informed Kernel to help guide efficient search for the best next action. Experimental results on Virtual Tools and PHYRE physical reasoning benchmarks show that Causal-PIK outperforms state-of-the-art results, requiring fewer actions to reach the goal. We also compare Causal-PIK to human studies, including results from a new user study we conducted on the PHYRE benchmark. We find that Causal-PIK remains competitive on tasks that are very challenging, even for human problem-solvers.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
