PulseAugur
EN
LIVE 08:34:15

New Bayesian Optimization Framework Enhances Bioprocess Development with Expert Input

Researchers have developed an enhanced Human-in-the-Loop Bayesian Optimization framework called Pareto Front Guided Sampling (PFGS). This framework allows domain experts to interactively select optimal candidates by reformulating Gaussian process surrogate-derived quantities into a multi-objective optimization problem. The system now incorporates constrained optimization by considering the probability of meeting specification limits and robust optimization by estimating performance degradation under input perturbations. The extended PFGS framework was demonstrated on a Chinese Hamster Ovary (CHO) cell culture simulator, successfully identifying operating conditions that are high-performing, feasible, and resilient to perturbations. AI

IMPACT This framework could improve efficiency and success rates in complex bioprocess development by integrating expert knowledge with advanced optimization techniques.

RANK_REASON The cluster describes a novel research paper detailing a new framework for optimization.

Read on arXiv stat.ML →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.LG TIER_1 English(EN) · Samuel Stricker, Claus Wirnsperger, Alessandro Butt\'e, Laura Helleckes, Gonzalo Guill\'en Gos\'albez, Antonio del Rio Chanona, Mehmet Mercang\"oz ·

    A Human-in-the-Loop Bayesian Optimization Framework for Constraint-Aware Bioprocess Development

    arXiv:2606.19230v1 Announce Type: new Abstract: This work presents an extension to Pareto Front Guided Sampling (PFGS), a Human-in-the-Loop (HitL) Bayesian Optimization (BO) framework in which Gaussian process (GP) surrogate-derived quantities are reformulated as objectives of a …

  2. arXiv stat.ML TIER_1 English(EN) · Mehmet Mercangöz ·

    A Human-in-the-Loop Bayesian Optimization Framework for Constraint-Aware Bioprocess Development

    This work presents an extension to Pareto Front Guided Sampling (PFGS), a Human-in-the-Loop (HitL) Bayesian Optimization (BO) framework in which Gaussian process (GP) surrogate-derived quantities are reformulated as objectives of a multi-objective optimization problem, and the re…