A Hierarchical Surrogate Model for Efficient Multi-Task Parameter Learning in Closed-Loop Contro

2025-08-18

AI Summary

To address the challenges of frequent controller parameter tuning, low data efficiency, and poor cross-task adaptability in multi-task closed-loop control, this paper proposes a structure-aware hierarchical Bayesian optimization (BO) framework. The method employs a hierarchical Gaussian process surrogate model that explicitly encodes structural priors from system dynamics, control laws, and closed-loop cost functions; it further enables exact propagation of task weights and cumulative costs via closed-form derivations. Theoretically, the approach guarantees sublinear regret while supporting both multi-task learning and knowledge transfer. In model predictive control simulations, it achieves significantly higher sample efficiency and markedly improved cross-task generalization compared to standard black-box BO. The core contribution lies in embedding domain-specific control structure knowledge into a hierarchical BO framework, thereby enabling efficient, interpretable, and transferable controller parameter learning.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Many control problems require repeated tuning and adaptation of controllers across distinct closed-loop tasks, where data efficiency and adaptability are critical. We propose a hierarchical Bayesian optimization (BO) framework that is tailored to efficient controller parameter learning in sequential decision-making and control scenarios for distinct tasks. Instead of treating the closed-loop cost as a black-box, our method exploits structural knowledge of the underlying problem, consisting of a dynamical system, a control law, and an associated closed-loop cost function. We construct a hierarchical surrogate model using Gaussian processes that capture the closed-loop state evolution under different parameterizations, while the task-specific weighting and accumulation into the closed-loop cost are computed exactly via known closed-form expressions. This allows knowledge transfer and enhanced data efficiency between different closed-loop tasks. The proposed framework retains sublinear regret guarantees on par with standard black-box BO, while enabling multi-task or transfer learning. Simulation experiments with model predictive control demonstrate substantial benefits in both sample efficiency and adaptability when compared to purely black-box BO approaches.

Problem

• Efficient controller tuning for multi-task closed-loop control
• Hierarchical Bayesian optimization for adaptable parameter learning
• Knowledge transfer between tasks using structured surrogate models

Innovation

• Hierarchical Bayesian optimization for controller tuning
• Gaussian processes model closed-loop state evolution
• Exact computation of task-specific cost functions