paper-with-me

Papers

Scalable Hyperparameter Optimization with Products of Gaussian Process Experts

2016-09-04 · ECML PKDD 2016 2016 9 · Nicolas Schilling, Martin Wistuba, Lars Schmidt-Thieme

In machine learning, hyperparameter optimization is a challenging but necessary task that is usually approached in a computationally expensive manner such as grid-search. Out of this reason, surrogate based black-box optimization techniques such as sequential model-based optimization have been proposed which allow for a faster hyperparameter optimization. Recent research proposes to also integrate hyperparameter performances on past data sets to allow for a faster and more efficient hyperparameter optimization. In this paper, we use products of Gaussian process experts as surrogate models for hyperparameter optimization. Naturally, Gaussian processes are a decent choice as they offer good prediction accuracy as well as estimations about their uncertainty. Additionally, their hyperparameters can be tuned very effectively. However, in the light of large meta data sets, learning a single Gaussian process is not feasible as it involves inversion of a large kernel matrix. This directly limits their usefulness for hyperparameter optimization if large scale hyperparameter performances on past data sets are given. By using products of Gaussian process experts the scalability issues can be circumvented, however, this usually comes with the price of having less predictive accuracy. In our experiments, we show empirically that products of experts nevertheless perform very well compared to a variety of published surrogate models. Thus, we propose a surrogate model that performs as well as the current state of the art, is scalable to large scale meta knowledge, does not include hyperparameters itself and finally is even very easy to parallelize.

📄 PDF Abstract BibTeX

Code (1)

nicoschilling/ECML2016

Tasks

Gaussian ProcessesHyperparameter Optimization

Methods 이 논문이 사용한 방법론

Gaussian Process Gaussian Processes are non-parametric models for approximating functions. They rely upon a measure of similarity between points (the kernel function) to predict the value for…

Similar Papers 제목 키워드 기반

Scalable Gaussian Process Hyperparameter Optimization via Coverage Regularization

2022-09-22 · Killian Wood, Alec M. Dunton, Amanda Muyskens, Benjamin W. Priest

Gaussian processes (GPs) are Bayesian non-parametric models popular in a variety of applications due to their accuracy and native uncertainty quantification (UQ). Tuning GP hyperparameters is critical to ensure the valid…

Conformal PredictionGaussian ProcessesHyperparameter OptimizationPrediction+1

Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)

2018-07-01 · ICML 2018 7 · Trefor Evans, Prasanth Nair

We introduce a kernel approximation strategy that enables computation of the Gaussian process log marginal likelihood and all hyperparameter derivatives in O(p) time. Our GRIEF kernel consists of p eigenfunctions fo…

Bayesian InferenceGaussian Processes

Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)

2018-07-05 · ICML 2018 · Trefor W. Evans, Prasanth B. Nair

We introduce a kernel approximation strategy that enables computation of the Gaussian process log marginal likelihood and all hyperparameter derivatives in $\mathcal{O}(p)$ time. Our GRIEF kernel consists of $p$ eigenfun…

Bayesian InferenceGaussian Processes

Preconditioning for Scalable Gaussian Process Hyperparameter Optimization

2021-07-01 · Jonathan Wenger, Geoff Pleiss, Philipp Hennig, John P. Cunningham 외

Gaussian process hyperparameter optimization requires linear solves with, and log-determinants of, large kernel matrices. Iterative numerical techniques are becoming popular to scale to larger datasets, relying on the co…

Gaussian ProcessesHyperparameter Optimization

Bias-Free Scalable Gaussian Processes via Randomized Truncations

2021-02-12 · Andres Potapczynski, Luhuan Wu, Dan Biderman, Geoff Pleiss 외

Scalable Gaussian Process methods are computationally attractive, yet introduce modeling biases that require rigorous study. This paper analyzes two common techniques: early truncated conjugate gradients (CG) and random …

Gaussian Processes