Learning to reason about and to act on physical cascading events

TL;DR

This paper introduces a supervised learning framework called Cascade, enabling agents to reason about and intervene in complex physical scenes with cascading events, using semantic tree search and event-driven models.

Contribution

It presents a novel setup for training agents to manipulate dynamic environments with cascades, combining semantic tree search and event-driven models for effective reasoning.

Findings

Agents successfully intervene in unseen scenes.

The approach reasons about alternative outcomes.

Effective in complex, non-linear cascades.

Abstract

Reasoning and interacting with dynamic environments is a fundamental problem in AI, but it becomes extremely challenging when actions can trigger cascades of cross-dependent events. We introduce a new supervised learning setup called {\em Cascade} where an agent is shown a video of a physically simulated dynamic scene, and is asked to intervene and trigger a cascade of events, such that the system reaches a "counterfactual" goal. For instance, the agent may be asked to "Make the blue ball hit the red one, by pushing the green ball". The agent intervention is drawn from a continuous space, and cascades of events makes the dynamics highly non-linear. We combine semantic tree search with an event-driven forward model and devise an algorithm that learns to search in semantic trees in continuous spaces. We demonstrate that our approach learns to effectively follow instructions to intervene in previously unseen complex scenes. It can also reason about alternative outcomes, when provided an observed cascade of events.