• français
    • English
  • English 
    • français
    • English
  • Login
JavaScript is disabled for your browser. Some features of this site may not work without it.
BIRD Home

Browse

This CollectionBy Issue DateAuthorsTitlesSubjectsJournals BIRDResearch centres & CollectionsBy Issue DateAuthorsTitlesSubjectsJournals

My Account

Login

Statistics

View Usage Statistics

Structured adaptive and random spinners for fast machine learning computations

Thumbnail
Date
2017
Notes
JMLR: W&CP, vol. 54
Link to item file
http://proceedings.mlr.press/v54/
Dewey
Intelligence artificielle
Sujet
Machine learning; random projections
Conference name
20th International Conference on Artificial Intelligence and Statistics (AISTATS 2017)
Conference date
04-2017
Conference city
Fort Lauderdale, FL
Conference country
United States
Book title
Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS 2017)
Author
Singh, Aarti; Zhu, Jerry
Publisher
JMLR: Workshop and Conference Proceedings
Year
2017
Pages number
1568
URI
https://basepub.dauphine.fr/handle/123456789/16505
Collections
  • LAMSADE : Publications
Metadata
Show full item record
Author
Bojarski, Mariusz
12319 NVIDIA [NVIDIA]
Choromanska, Anna
4768 Department of Electrical and Computer Engineering
128119 Polytechnic institute of New York University [NYU-Poly]
Choromanski, Krzysztof
status unknown
Fagan, Francois
172051 Industrial Engineering and Operations Research Department [IEOR Dept]
Gouy-Pailler, Cédric
40217 Laboratoire d'Intégration des Systèmes et des Technologies [LIST]
Morvan, Anne
989 Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Sakr, Nourhan
172051 Industrial Engineering and Operations Research Department [IEOR Dept]
Sarlos, Tamas
status unknown
Atif, Jamal
989 Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Type
Communication / Conférence
Item number of pages
1020-1029
Abstract (EN)
We consider an efficient computational framework for speeding up several machine learning algorithms with almost no loss of accuracy. The proposed framework relies on projections via structured matrices that we call Structured Spinners, which are formed as products of three structured matrix-blocks that incorporate rotations. The approach is highly generic, i.e. i) structured matrices under consideration can either be fully-randomized or learned, ii) our structured family contains as special cases all previously considered structured schemes, iii) the setting extends to the non-linear case where the projections are followed by non-linear functions, and iv) the method finds numerous applications including kernel approximations via random feature maps, dimensionality reduction algorithms, new fast cross-polytope LSH techniques, deep learning, convex optimization algorithms via Newton sketches, quantization with random projection trees, and more. The proposed framework comes with theoretical guarantees characterizing the capacity of the structured model in reference to its unstructured counterpart and is based on a general theoretical principle that we describe in the paper. As a consequence of our theoretical analysis, we provide the first theoretical guarantees for one of the most efficient existing LSH algorithms based on the HD3HD2HD1 structured matrix [Andoni et al., 2015]. The exhaustive experimental evaluation confirms the accuracy and efficiency of structured spinners for a variety of different applications.

  • Accueil Bibliothèque
  • Site de l'Université Paris-Dauphine
  • Contact
SCD Paris Dauphine - Place du Maréchal de Lattre de Tassigny 75775 Paris Cedex 16

 Content on this site is licensed under a Creative Commons 2.0 France (CC BY-NC-ND 2.0) license.