Increasing the Scope as You Learn: Adaptive Bayesian Optimization in Nested Subspaces
Papenmeier, Leonard; Nardi, Luigi; Poloczek, Matthias
Zusammenfassung
Recent advances have extended the scope of Bayesian optimization (BO) to expensive-to-evaluate black-box functions with dozens of dimensions, aspiring to unlock impactful applications, for example, in the life sciences, neural architecture search, and robotics. However, a closer examination reveals that the state-of-the-art methods for high-dimensional Bayesian optimization (HDBO) suffer from degrading performance as the number of dimensions increases, or even risk failure if certain unverifiable assumptions are not met. This paper proposes BAxUS that leverages a novel family of nested random subspaces to adapt the space it optimizes over to the problem. This ensures high performance while removing the risk of failure, which we assert via theoretical guarantees. A comprehensive evaluation demonstrates that BAxUS achieves better results than the state-of-the-art methods for a broad set of applications.
Schlüsselwörter
Bayesian Optimization
Zitieren als
Papenmeier, L., Nardi, L., & Poloczek, M. (2022). Increasing the Scope as You Learn: Adaptive Bayesian Optimization in Nested Subspaces. In Koyejo, S., Mohamed, S., Agarwal, A., Belgrave, D., Cho, K., & Oh, A. (Eds.),
Advances in Neural Information Processing Systems 35 (NeurIPS 2022) (-, pp. 11586–11601). Advances in Neural Information Processing Systems: Vol. 35. Red Hook, NY, United States: Curran Associates.
Details
Publikationstyp
Forschungsartikel in Sammelband (Konferenz)
Begutachtet
Ja
Publikationsstatus
Veröffentlicht
Jahr
2022
Konferenz
Advances in Neural Information Processing Systems 35, NeurIPS 2022
Konferenzort
New Orleans Convention Center
Band
35
Buchtitel
Advances in Neural Information Processing Systems 35 (NeurIPS 2022)
Herausgeber
Koyejo, S.; Mohamed, S.; Agarwal, A.; Belgrave, D.; Cho, K.; Oh, A.
Erste Seite
11586
Letzte Seite
11601
Auflage
-
Band
35
Reihe
Advances in Neural Information Processing Systems
Verlag
Curran Associates
Ort
Red Hook, NY, United States
Sprache
Englisch
ISSN
1049-5258
ISBN
9781713871088
Gesamter Text
