Matches in SemOpenAlex for { <https://semopenalex.org/work/W3211464761> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W3211464761 abstract "Optimizing multiple competing black-box objectives is a challenging problem in many fields, including science, engineering, and machine learning. Multi-objective Bayesian optimization is a powerful approach for identifying the optimal trade-offs between the objectives with very few function evaluations. However, existing methods tend to perform poorly when observations are corrupted by noise, as they do not take into account uncertainty in the true Pareto frontier over the previously evaluated designs. We propose a novel acquisition function, NEHVI, that overcomes this important practical limitation by applying a Bayesian treatment to the popular expected hypervolume improvement criterion to integrate over this uncertainty in the Pareto frontier. We further argue that, even in the noiseless setting, the problem of generating multiple candidates in parallel reduces that of handling uncertainty in the Pareto frontier. Through this lens, we derive a natural parallel variant of NEHVI that can efficiently generate large batches of candidates. We provide a theoretical convergence guarantee for optimizing a Monte Carlo estimator of NEHVI using exact sample-path gradients. Empirically, we show that NEHVI achieves state-of-the-art performance in noisy and large-batch environments." @default.
- W3211464761 created "2021-11-22" @default.
- W3211464761 creator A5006700143 @default.
- W3211464761 creator A5055451857 @default.
- W3211464761 creator A5075229217 @default.
- W3211464761 date "2021-12-06" @default.
- W3211464761 modified "2023-09-29" @default.
- W3211464761 title "Parallel Bayesian Optimization of Multiple Noisy Objectives with Expected Hypervolume Improvement" @default.
- W3211464761 hasPublicationYear "2021" @default.
- W3211464761 type Work @default.
- W3211464761 sameAs 3211464761 @default.
- W3211464761 citedByCount "0" @default.
- W3211464761 crossrefType "proceedings-article" @default.
- W3211464761 hasAuthorship W3211464761A5006700143 @default.
- W3211464761 hasAuthorship W3211464761A5055451857 @default.
- W3211464761 hasAuthorship W3211464761A5075229217 @default.
- W3211464761 hasConcept C105795698 @default.
- W3211464761 hasConcept C107673813 @default.
- W3211464761 hasConcept C119857082 @default.
- W3211464761 hasConcept C126255220 @default.
- W3211464761 hasConcept C137635306 @default.
- W3211464761 hasConcept C154945302 @default.
- W3211464761 hasConcept C162324750 @default.
- W3211464761 hasConcept C185429906 @default.
- W3211464761 hasConcept C19499675 @default.
- W3211464761 hasConcept C2777303404 @default.
- W3211464761 hasConcept C2778049539 @default.
- W3211464761 hasConcept C33923547 @default.
- W3211464761 hasConcept C41008148 @default.
- W3211464761 hasConcept C50522688 @default.
- W3211464761 hasConcept C94966114 @default.
- W3211464761 hasConceptScore W3211464761C105795698 @default.
- W3211464761 hasConceptScore W3211464761C107673813 @default.
- W3211464761 hasConceptScore W3211464761C119857082 @default.
- W3211464761 hasConceptScore W3211464761C126255220 @default.
- W3211464761 hasConceptScore W3211464761C137635306 @default.
- W3211464761 hasConceptScore W3211464761C154945302 @default.
- W3211464761 hasConceptScore W3211464761C162324750 @default.
- W3211464761 hasConceptScore W3211464761C185429906 @default.
- W3211464761 hasConceptScore W3211464761C19499675 @default.
- W3211464761 hasConceptScore W3211464761C2777303404 @default.
- W3211464761 hasConceptScore W3211464761C2778049539 @default.
- W3211464761 hasConceptScore W3211464761C33923547 @default.
- W3211464761 hasConceptScore W3211464761C41008148 @default.
- W3211464761 hasConceptScore W3211464761C50522688 @default.
- W3211464761 hasConceptScore W3211464761C94966114 @default.
- W3211464761 hasLocation W32114647611 @default.
- W3211464761 hasOpenAccess W3211464761 @default.
- W3211464761 hasPrimaryLocation W32114647611 @default.
- W3211464761 hasRelatedWork W2037214880 @default.
- W3211464761 hasRelatedWork W2090523679 @default.
- W3211464761 hasRelatedWork W2137432375 @default.
- W3211464761 hasRelatedWork W2402456051 @default.
- W3211464761 hasRelatedWork W2554871335 @default.
- W3211464761 hasRelatedWork W2562293769 @default.
- W3211464761 hasRelatedWork W2896993302 @default.
- W3211464761 hasRelatedWork W2902095000 @default.
- W3211464761 hasRelatedWork W2950853128 @default.
- W3211464761 hasRelatedWork W2964664157 @default.
- W3211464761 hasRelatedWork W2998166751 @default.
- W3211464761 hasRelatedWork W3035058078 @default.
- W3211464761 hasRelatedWork W3035677605 @default.
- W3211464761 hasRelatedWork W3158465438 @default.
- W3211464761 hasRelatedWork W3160947956 @default.
- W3211464761 hasRelatedWork W3162387221 @default.
- W3211464761 hasRelatedWork W3164564312 @default.
- W3211464761 hasRelatedWork W3175328119 @default.
- W3211464761 hasRelatedWork W3203318300 @default.
- W3211464761 hasRelatedWork W53735871 @default.
- W3211464761 hasVolume "34" @default.
- W3211464761 isParatext "false" @default.
- W3211464761 isRetracted "false" @default.
- W3211464761 magId "3211464761" @default.
- W3211464761 workType "article" @default.