arXiv AI recent: Causal Object-Centric Models for Planning with Monte Carlo Tree Search
Researchers introduced COMET, a model-based reinforcement learning algorithm that performs Monte Carlo Tree Search in a slot-structured latent space.,COMET pairs a frozen unsupervised obj...
COMET uses a novel action-slot fusion mechanism and object-causal attention to modulate token interactions and concentrate decision-making on task-relevant entities.,The algorithm was tested on tasks from the Object-Centric Visual RL benchmark, ManiSkill, Robosuite, and VizDoom, and outperformed...