Practical Transfer Learning for Bayesian Optimization Article Swipe

PDF

Related Concepts

Bayesian optimization Hyperparameter Computer science Hyperparameter optimization Bayesian probability Machine learning Artificial intelligence Gaussian process Benchmark (surveying) Transfer of learning Optimization problem Mathematical optimization Gaussian Algorithm Mathematics Support vector machine Quantum mechanics Geography Geodesy Physics

Matthias Feurer , Benjamin Letham , Frank Hutter , Eytan Bakshy ·

YOU? · · 2018 · Open Access · · DOI: https://doi.org/10.48550/arxiv.1802.02219 · OA: W3159088669

When hyperparameter optimization of a machine learning algorithm is repeated for multiple datasets it is possible to transfer knowledge to an optimization run on a new dataset. We develop a new hyperparameter-free ensemble model for Bayesian optimization that is a generalization of two existing transfer learning extensions to Bayesian optimization and establish a worst-case bound compared to vanilla Bayesian optimization. Using a large collection of hyperparameter optimization benchmark problems, we demonstrate that our contributions substantially reduce optimization time compared to standard Gaussian process-based Bayesian optimization and improve over the current state-of-the-art for transfer hyperparameter optimization.