Pragmatic Curiosity: A Unified Framework for Hybrid Learning and Optimization via Active Inference

TL;DR

Pragmatic Curiosity (PraC) is a unified active inference framework that balances information gain and regret to improve hybrid learning and optimization tasks involving expensive black-box evaluations.

Contribution

It introduces a novel, flexible approach that integrates goal-directed optimization and information-seeking in a single framework, applicable across various complex regimes.

Findings

PraC reduces downstream decision risk.

It improves coverage of critical outcome regions.

It jointly learns predictive and preference structures without task-specific staging.

Abstract

Many engineering and scientific workflows rely on expensive black-box evaluations, requiring sequential decisions that must both improve task performance and reduce uncertainty. Bayesian optimization (BO) and Bayesian experimental design (BED) provide powerful but largely separate treatments of goal-directed optimization and information-seeking experimentation, leaving limited guidance for hybrid settings in which learning and optimization are intrinsically coupled. We propose Pragmatic Curiosity (PraC), a unified framework for hybrid learning and optimization via active inference. PraC evaluates candidate queries by trading information gain about a task-relevant latent symbol against an expected regret-based potential over outcomes. This formulation exposes three operational design choices: which latent quantity should be clarified, how task value is encoded as regret, and how strongly information gain should be exchanged against pragmatic value. We instantiate PraC across three regimes of increasing complexity: decision-oriented plume monitoring with fixed global symbols and known downstream losses, targeted active search with induced local symbols and evolving coverage goals, and composite Bayesian optimization with hierarchical regret learning under unknown preferences. Across these regimes, PraC reduces downstream decision risk, improves coverage of critical outcome regions, and jointly learns predictive and preference structures without relying on task-specific staging rules.