# An Information-theoretic On-line Learning Principle for Specialization in Hierarchical Decision-Making Systems

## TL;DR

This paper introduces an information-theoretic on-line learning principle that enables specialized decision-makers with limited information-processing capabilities to collaboratively solve complex decision problems by learning optimal problem partitions.

## Contribution

It proposes a novel on-line learning rule based on information-theoretic bounded rationality that drives division of labor and specialization in hierarchical decision-making systems.

## Key findings

- The approach effectively partitions complex problems for specialized linear policies.
- The method applies to classification, regression, reinforcement learning, and adaptive control.
- It demonstrates improved decision-making in resource-limited, complex environments.

## Abstract

Information-theoretic bounded rationality describes utility-optimizing decision-makers whose limited information-processing capabilities are formalized by information constraints. One of the consequences of bounded rationality is that resource-limited decision-makers can join together to solve decision-making problems that are beyond the capabilities of each individual. Here, we study an information-theoretic principle that drives division of labor and specialization when decision-makers with information constraints are joined together. We devise an on-line learning rule of this principle that learns a partitioning of the problem space such that it can be solved by specialized linear policies. We demonstrate the approach for decision-making problems whose complexity exceeds the capabilities of individual decision-makers, but can be solved by combining the decision-makers optimally. The strength of the model is that it is abstract and principled, yet has direct applications in classification, regression, reinforcement learning and adaptive control.