7 papers
SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization
Leonard Papenmeier, Petru Tighineanu
Multi-objective optimization aims to solve problems with competing objectives. Evaluating such problems is often slow or expensive, limiting the budget of evaluations. In many appl…
Benchmarking that Matters: Rethinking Benchmarking for Practical Impact
Anna V. Kononova, Niki van Stein, Olaf Mersmann +14
Benchmarking has driven scientific progress in Evolutionary Computation, yet current practices fall short of real-world needs. Widely used synthetic suites such as BBOB and CEC iso…
Understanding High-Dimensional Bayesian Optimization
Leonard Papenmeier, Matthias Poloczek, Luigi Nardi
Recent work reported that simple Bayesian optimization (BO) methods perform well for high-dimensional real-world tasks, seemingly contradicting prior work and tribal knowledge. Thi…
A Unified Framework for Entropy Search and Expected Improvement in Bayesian Optimization
Nuojin Cheng, Leonard Papenmeier, Stephen Becker +1
Bayesian optimization is a widely used method for optimizing expensive black-box functions, with Expected Improvement being one of the most commonly used acquisition functions. In…
Bencher: Simple and Reproducible Benchmarking for Black-Box Optimization
Leonard Papenmeier, Luigi Nardi
We present Bencher, a modular benchmarking framework for black-box optimization that fundamentally decouples benchmark execution from optimization logic. Unlike prior suites that f…
Leveraging Axis-Aligned Subspaces for High-Dimensional Bayesian Optimization with Group Testing
Erik Hellsten, Carl Hvarfner, Leonard Papenmeier +1
Bayesian optimization (BO ) is an effective method for optimizing expensive-to-evaluate black-box functions. While high-dimensional problems can be particularly challenging, due to…