human-in-the-loop bayesian optimization

This study addresses the challenge in bioprocess development of simultaneously optimizing performance, constraint satisfaction, and operational robustness—a task traditionally reliant on expert judgment. To this end, the authors propose a human-in-the-loop multi-objective Bayesian optimization framework that explicitly incorporates the probability of constraint satisfaction and robustness under input perturbations into the Pareto optimization objectives. The approach integrates Gaussian process surrogate models, Monte Carlo–based robustness evaluation, and Pareto-guided sampling, complemented by an interactive four-dimensional visualization interface to support dynamic expert decision-making. Demonstrated on an eight-dimensional fed-batch CHO cell culture simulation, the method efficiently identifies high-performing, feasible, and robust operating conditions, substantially improving experimental resource efficiency and enabling more intelligent termination criteria for iterative optimization.

This work addresses the lack of convergence guarantees for non-Gaussian process (non-GP) surrogate models in Bayesian optimization (BO). To resolve this, we propose the first axiomatic pseudo-BO framework, formally characterizing the minimal conditions required for sequential black-box optimization to converge. Methodologically, we design a lightweight local-regression-based surrogate model coupled with a randomized prior mechanism for efficient uncertainty quantification, and integrate it with upper-confidence-bound-type acquisition strategies. Theoretically, we provide the first rigorous convergence analysis for non-GP BO. Empirically, our framework consistently outperforms state-of-the-art methods—including GP-BO, TuRBO, and ALEBO—across high-dimensional synthetic benchmarks, neural network hyperparameter tuning, and robot control tasks. Thus, it achieves both theoretical soundness and practical superiority.

Efficiently identifying the Pareto-optimal set for multi-objective black-box functions over high-dimensional continuous design spaces remains challenging due to the exponential growth of candidate solutions. Method: This paper proposes Adaptive ε-PAL, the first algorithm integrating tree-based adaptive discretization with Gaussian process (GP) Bayesian optimization. It leverages GP surrogate modeling, Pareto front estimation, and information-theoretic analysis of compact metric spaces to dynamically partition the input space—thereby avoiding exhaustive enumeration. Contribution/Results: We establish a provably tight upper bound on the ε-accurate sample complexity. Empirical evaluation across multiple benchmarks demonstrates substantial reductions in function evaluations compared to state-of-the-art Pareto-set identification methods. The core contribution is a theoretically grounded, adaptive multi-objective optimization framework that simultaneously ensures rigorous convergence guarantees and practical efficiency.

To address the lack of online user intervention capability in Bayesian optimization (BO) for hyperparameter tuning, this paper proposes an intervenable BO framework supporting dynamic prior injection. Unlike existing approaches that only permit expert knowledge incorporation during initialization, our method enables multiple, real-time integrations of user preferences and domain knowledge as prior distributions throughout the optimization process, complemented by an anomaly-prior detection mechanism to ensure robustness. Built upon πBO, it introduces adaptive prior updating while preserving theoretical convergence guarantees, thereby significantly enhancing controllability and transparency without compromising optimization performance. Experiments demonstrate that the framework effectively accelerates convergence when beneficial priors are provided, reliably rejects misleading priors, and achieves performance on par with standard BO across multiple tasks.

Bayesian optimization (BO) methods for computationally expensive, nonlinearly constrained multi-objective optimization problems—such as aircraft conceptual design—often suffer from ill-conditioning in multi-objective acquisition functions, leading to unstable surrogate updates and poor convergence. Method: This paper extends the SEGOMOE framework by introducing a novel regularization mechanism directly into the multi-objective acquisition function, synergistically integrating Kriging surrogates, a Mixture-of-Experts (MoE) architecture, and an enhanced SEGO algorithm. Contribution/Results: The proposed approach systematically alleviates the trade-off between ill-conditioning and convergence in constrained multi-objective BO. Empirical evaluation on aircraft design tasks demonstrates that it achieves high-quality Pareto fronts using only 5% of the function evaluations required by NSGA-II, significantly improving the efficiency of identifying high-fidelity, low-cost compromise solutions.

This work presents a general-purpose desktop platform based on Bayesian optimization to address both single- and multi-objective optimization problems commonly encountered in scientific experimentation and industrial processes. The platform features a graphical user interface that accepts user-defined input parameters, optimization objectives, and historical data, and integrates surrogate models—such as Gaussian processes—with customizable acquisition functions tailored for single- or multi-objective settings to automatically guide iterative experimentation. Its key innovation lies in encapsulating advanced Bayesian optimization techniques into an accessible, configurable, and interactive tool that significantly reduces the number of required experiments while efficiently converging toward optimal solutions, thereby enhancing experimental efficiency and the intelligence of decision-making processes.

This study addresses the challenge that the performance of Bayesian optimization (BO) heavily depends on hyperparameter presets by proposing a data-driven bilevel BO framework for automatic parameter tuning. Methodologically, it introduces a novel pretraining paradigm that infers Gaussian processes from initial observations and generates sample paths. The outer level employs cumulative regret as the evaluation metric to automatically search for optimal hyperparameter configurations via Bayesian optimization. Experimental results demonstrate that this bilevel architecture efficiently identifies highly robust hyperparameter combinations from the candidate space, significantly reducing the cost of manual tuning while enhancing overall optimization efficiency.

本文提出了一种基于产品专家高斯过程模型（GP-pro-c）的贝叶斯优化算法BO-pro-c，解决了大规模优化问题中的计算复杂度限制，并通过实验验证了其在保持优化性能的同时降低了简单遗憾和计算开销。

This study addresses the low sample efficiency caused by coupled evaluations in multi-objective Bayesian optimization by proposing a correlation-based decoupled evaluation strategy. Methodologically, it jointly models a multi-task Gaussian process and designs a total correlation metric to dynamically select optimal evaluation subsets, exploiting intrinsic correlations between objectives and constraints to enable decoupled evaluations. Theoretically, the asymptotic consistency of this strategy is established, along with its advantage in maximizing posterior entropy reduction for unevaluated tasks. Experimental results demonstrate that the proposed method significantly outperforms existing coupled and decoupled baselines, effectively enhancing optimization performance, sample efficiency, and convergence speed while reducing computational costs.

This study addresses the impracticality of fully Bayesian approaches in black-box optimization, where computationally expensive Markov chain Monte Carlo (MCMC) sampling induces prohibitive decision latency. We propose ELF-BO, an algorithm that introduces a novel "sample-while-evaluating" asynchronous mechanism. By exploiting the idle time during objective function evaluations to sample the hyperparameter posterior distribution in parallel, and by incorporating an importance reweighting strategy, ELF-BO entirely conceals the costly MCMC computations within the evaluation latency, thereby eliminating additional decision overhead. Experiments on both synthetic and real-world tasks demonstrate that the proposed method achieves performance comparable to full Bayesian optimization while exhibiting lower decision latency than standard Bayesian optimization. This work renders fully Bayesian optimization practically viable for the first time.