A Variational Inequality Model for Learning Neural Networks

Patrick Louis Combettes, Jean-Christophe Pesquet, Audrey Repetti

Research output: Chapter in Book/Report/Conference proceedingConference contribution

34 Downloads (Pure)

Abstract

Neural networks have become ubiquitous tools for solving signal and image processing problems, and they often outperform standard approaches. Nevertheless, training the layers of a neural network is a challenging task in many applications. The prevalent training procedure consists of minimizing highly non-convex objectives based on data sets of huge dimension. In this context, current methodologies are not guaranteed to produce global solutions. We present an alternative approach which foregoes the optimization framework and adopts a variational inequality formalism. The associated algorithm guarantees convergence of the iterates to a true solution of the variational inequality and it possesses an efficient block-iterative structure. A numerical application is presented.
Original languageEnglish
Title of host publication2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
PublisherIEEE
ISBN (Electronic)9781728163277
DOIs
Publication statusPublished - 5 May 2023
Event48th IEEE International Conference on Acoustics, Speech and Signal Processing 2023 - Rhodes Island, Rhodes Island, Greece
Duration: 4 Jun 202310 Jun 2023
Conference number: 48
https://2023.ieeeicassp.org/
https://2023.ieeeicassp.org

Conference

Conference48th IEEE International Conference on Acoustics, Speech and Signal Processing 2023
Abbreviated titleICASSP 2023
Country/TerritoryGreece
CityRhodes Island
Period4/06/2310/06/23
Internet address

Keywords

  • Block-iterative algorithm
  • MRI
  • neural networks
  • transfer learning
  • variational inequality

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'A Variational Inequality Model for Learning Neural Networks'. Together they form a unique fingerprint.

Cite this