Action And Perception As Divergence Minimization

Abstract

To learn directed behaviors in complex environments, intelligent agents need to optimize objective functions. Various objectives are known for designing artificial agents, including task rewards and intrinsic motivation. However, it is unclear how the known objectives relate to each other, which objectives remain yet to be discovered, and which objectives better describe the behavior of humans. We introduce the Action Perception Divergence (APD), an approach for categorizing the space of possible objective functions for embodied agents. We show a spectrum that reaches from narrow to general objectives. While the narrow objectives correspond to domain-specific rewards as typical in reinforcement learning, the general objectives maximize information with the environment through latent variable models of input sequences. Intuitively, these agents use perception to align their beliefs with the world and use actions to align the world with their beliefs. They infer representations that are

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyhafner2020action

Related papers

• Understanding The Origin Of Information-seeking Exploration In Probabilistic Objectives For Control (2021)0.00
• Expanding The Active Inference Landscape: More Intrinsic Motivations In The Perception-action Loop (2018)9.92
• Information Is Power: Intrinsic Control Via Information Capture (2021)0.00
• Goal-oriented Inference Of Environment From Redundant Observations (2023)3.58
• Where Do You Think You're Going?: Inferring Beliefs About Dynamics From Behavior (2018)0.00
• Towards Measuring Goal-directedness In AI Systems (2024)0.00
• Evaluating Agents Without Rewards (2020)0.00
• GRASP: Gradient Realignment Via Active Shared Perception For Multi-agent Collaborative Optimization (2026)0.00