A Unified Bellman Optimality Principle Combining Reward Maximization And Empowerment

Abstract

Empowerment is an information-theoretic method that can be used to intrinsically motivate learning agents. It attempts to maximize an agent's control over the environment by encouraging visiting states with a large number of reachable next states. Empowered learning has been shown to lead to complex behaviors, without requiring an explicit reward signal. In this paper, we investigate the use of empowerment in the presence of an extrinsic reward signal. We hypothesize that empowerment can guide reinforcement learning (RL) agents to find good early behavioral solutions by encouraging highly empowered states. We propose a unified Bellman optimality principle for empowered reward maximization. Our empowered reward maximization approach generalizes both Bellman's optimality principle as well as recent information-theoretical extensions to it. We prove uniqueness of the empowered values and show convergence to the optimal solution. We then apply this idea to develop off-policy actor-critic R

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyleibfried2019a

Related papers

• Experimental Evidence That Empowerment May Drive Exploration In Sparse-reward Environments (2021)0.00
• A Unified Strategy For Implementing Curiosity And Empowerment Driven Reinforcement Learning (2018)0.00
• Unsupervised Real-time Control Through Variational Empowerment (2017)3.58
• Variational Empowerment As Representation Learning For Goal-based Reinforcement Learning (2021)0.00
• Towards Empowerment Gain Through Causal Structure Learning In Model-based RL (2025)0.00
• Robust Multi-agent Reinforcement Learning With Social Empowerment For Coordination And Communication (2020)0.00
• Learning Efficient Representation For Intrinsic Motivation (2019)0.00
• Reliably Re-acting To Partner's Actions With The Social Intrinsic Motivation Of Transfer Empowerment (2022)0.00