Online function minimization with convex random relu expansions

Laurens Bliek, Michel Verhaegen, Sander Wahls

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

1 Citaat (Scopus)

Samenvatting

We propose CDONE, a convex version of the DONE algorithm. DONE is a derivative-free online optimization algorithm that uses surrogate modeling with noisy measurements to find a minimum of objective functions that are expensive to evaluate. Inspired by their success in deep learning, CDONE makes use of rectified linear units, together with a nonnegativity constraint to enforce convexity of the surrogate model. This leads to a sparse and cheap to evaluate surrogate model of the unknown optimization objective that is still accurate and that can be minimized with convex optimization algorithms. The CDONE algorithm is demonstrated on a toy example and on the problem of hyper-parameter optimization for a deep learning example on handwritten digit classification.

Originele taal-2Engels
Titel2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 - Proceedings
RedacteurenNaonori Ueda, Jen-Tzung Chien, Tomoko Matsui, Jan Larsen, Shinji Watanabe
UitgeverijIEEE Computer Society
Pagina's1-6
Aantal pagina's6
ISBN van elektronische versie9781509063413
DOI's
StatusGepubliceerd - 5 dec 2017
Extern gepubliceerdJa
Evenement2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 - Tokyo, Japan
Duur: 25 sep 201728 sep 2017

Congres

Congres2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017
LandJapan
StadTokyo
Periode25/09/1728/09/17

Vingerafdruk Duik in de onderzoeksthema's van 'Online function minimization with convex random relu expansions'. Samen vormen ze een unieke vingerafdruk.

Citeer dit