Adaptive multi-fidelity optimization with ...
Document type :
Communication dans un congrès avec actes
Title :
Adaptive multi-fidelity optimization with fast learning rates
Author(s) :
Fiegel, Côme [Auteur]
Scool [Scool]
Département d'informatique - ENS-PSL [DI-ENS]
Gabillon, Victor [Auteur]
Huawei R&D [United Kingdom]
Valko, Michal [Auteur]
Scool [Scool]
Département d'informatique - ENS-PSL [DI-ENS]
Gabillon, Victor [Auteur]
Huawei R&D [United Kingdom]
Valko, Michal [Auteur]

Conference title :
International Conference on Artificial Intelligence and Statistics
City :
Palermo
Country :
Italie
Start date of the conference :
2020
HAL domain(s) :
Statistiques [stat]/Machine Learning [stat.ML]
English abstract : [en]
In multi-fidelity optimization, we have access to biased approximations of varying costs of the target function. In this work, we study the setting of optimizing a locally smooth function with a limited budget Λ, where the ...
Show more >In multi-fidelity optimization, we have access to biased approximations of varying costs of the target function. In this work, we study the setting of optimizing a locally smooth function with a limited budget Λ, where the learner has to make a trade-off between the cost and the bias of these approximations. We first prove lower bounds for the simple regret under different assumptions on the fidelities, based on a cost-to-bias function. We then present the Kometo algorithm which achieves, with additional logarithmic factors, the same rates without any knowledge of the function smoothness and fidelity assumptions and improving prior results. Finally, we empirically show that our algorithm outperforms prior multi-fidelity optimization methods without the knowledge of problem-dependent parameters.Show less >
Show more >In multi-fidelity optimization, we have access to biased approximations of varying costs of the target function. In this work, we study the setting of optimizing a locally smooth function with a limited budget Λ, where the learner has to make a trade-off between the cost and the bias of these approximations. We first prove lower bounds for the simple regret under different assumptions on the fidelities, based on a cost-to-bias function. We then present the Kometo algorithm which achieves, with additional logarithmic factors, the same rates without any knowledge of the function smoothness and fidelity assumptions and improving prior results. Finally, we empirically show that our algorithm outperforms prior multi-fidelity optimization methods without the knowledge of problem-dependent parameters.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.inria.fr/hal-03288879/document
- Open access
- Access the document
- https://hal.inria.fr/hal-03288879/document
- Open access
- Access the document
- https://hal.inria.fr/hal-03288879/document
- Open access
- Access the document
- document
- Open access
- Access the document
- fiegel2020adaptive.pdf
- Open access
- Access the document
- document
- Open access
- Access the document
- fiegel2020adaptive.pdf
- Open access
- Access the document