Pragmatic Curiosity: A Hybrid Learning-Optimization Paradigm via Active Inference
This work addresses the challenge of intrinsically coupled learning and optimization for engineers and scientists dealing with expensive black-box evaluations, offering a unified approach where previous methods were largely separate.
This paper introduces "pragmatic curiosity," a new paradigm for decision-making in expensive black-box evaluation workflows, which simultaneously optimizes performance and reduces uncertainty. Derived from active inference, this method consistently outperforms strong Bayesian optimization and Bayesian experimental design baselines across various hybrid tasks, achieving higher estimation accuracy, better critical-region coverage, and improved final solution quality.
Many engineering and scientific workflows depend on expensive black-box evaluations, requiring decision-making that simultaneously improves performance and reduces uncertainty. Bayesian optimization (BO) and Bayesian experimental design (BED) offer powerful yet largely separate treatments of goal-seeking and information-seeking, providing limited guidance for hybrid settings where learning and optimization are intrinsically coupled. We propose "pragmatic curiosity," a hybrid learning-optimization paradigm derived from active inference, in which actions are selected by minimizing the expected free energy--a single objective that couples pragmatic utility with epistemic information gain. We demonstrate the practical effectiveness and flexibility of pragmatic curiosity on various real-world hybrid tasks, including constrained system identification, targeted active search, and composite optimization with unknown preferences. Across these benchmarks, pragmatic curiosity consistently outperforms strong BO-type and BED-type baselines, delivering higher estimation accuracy, better critical-region coverage, and improved final solution quality.