A Unified Bellman Equation for Causal Information and Value in Markov Decision Processes
arXiv:1703.01585
Abstract
The interaction between an artificial agent and its environment is bi-directional. The agent extracts relevant information from the environment, and affects the environment by its actions in return to accumulate high expected reward. Standard reinforcement learning (RL) deals with the expected reward maximization. However, there are always information-theoretic limitations that restrict the expected reward, which are not properly considered by the standard RL. In this work we consider RL objectives with information-theoretic limitations. For the first time we derive a Bellman-type recursive equa- tion for the causal information between the environment and the agent, which is combined plausibly with the Bellman recursion for the value function. The unified equitation serves to explore the typical behavior of artificial agents in an infinite time horizon.
9 pages, 4 figures
References in corpus (3)
Cited by in corpus (10)
- A unified strategy for implementing curiosity and empowerment driven reinforcement learning
- Exploiting Hierarchy for Learning and Transfer in KL-regularized RL
- Dynamics Generalization via Information Bottleneck in Deep Reinforcement Learning
- A Unified Bellman Optimality Principle Combining Reward Maximization and Empowerment
- Behavior Priors for Efficient Reinforcement Learning
- Curiosity-driven reinforcement learning with homeostatic regulation
- A Tutorial on Sparse Gaussian Processes and Variational Inference
- The Journey is the Reward: Unsupervised Learning of Influential Trajectories
- An Information-Theoretic Perspective on Credit Assignment in Reinforcement Learning
- GEM: Group Enhanced Model for Learning Dynamical Control Systems