Robust Data-Driven Accelerated Mirror Descent

Hong Ye Tan; Subhadip Mukherjee; Junqi Tang; Andreas Hauptmann; Carola-Bibiane Schönlieb

doi:10.1109/ICASSP49357.2023.10096875

Robust Data-Driven Accelerated Mirror Descent

Hong Ye Tan, Subhadip Mukherjee, Junqi Tang, Andreas Hauptmann, Carola-Bibiane Schönlieb

Mathematics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

19 Downloads (Pure)

Abstract

Learning-to-optimize is an emerging framework that leverages training data to speed up the solution of certain optimization problems. One such approach is based on the classical mirror descent algorithm, where the mirror map is modelled using input-convex neural networks. In this work, we extend this functional parameterization approach by introducing momentum into the iterations, based on the classical accelerated mirror descent. Our approach combines short-time accelerated convergence with stable long-time behavior. We empirically demonstrate additional robustness with respect to multiple parameters on denoising and deconvolution experiments.

Original language	English
Title of host publication	ICASSP 2023 - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher	IEEE
ISBN (Electronic)	9781728163277
ISBN (Print)	9781728163284
DOIs	https://doi.org/10.1109/ICASSP49357.2023.10096875
Publication status	E-pub ahead of print - 5 May 2023
Event	2023 IEEE International Conference on Acoustics, Speech and Signal Processing - Rhodes Island, Greece Duration: 4 Jun 2023 → 10 Jun 2023

Conference

Conference	2023 IEEE International Conference on Acoustics, Speech and Signal Processing
Abbreviated title	ICASSP 2023
Country/Territory	Greece
Period	4/06/23 → 10/06/23

Access to Document

10.1109/ICASSP49357.2023.10096875Licence: None: All rights reserved

TanHY2023Robust
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript, 361 KBLicence: Other (please specify with Rights Statement)

Cite this

@inproceedings{9d89645dbec546b08dcbae0f78827280,

title = "Robust Data-Driven Accelerated Mirror Descent",

abstract = "Learning-to-optimize is an emerging framework that leverages training data to speed up the solution of certain optimization problems. One such approach is based on the classical mirror descent algorithm, where the mirror map is modelled using input-convex neural networks. In this work, we extend this functional parameterization approach by introducing momentum into the iterations, based on the classical accelerated mirror descent. Our approach combines short-time accelerated convergence with stable long-time behavior. We empirically demonstrate additional robustness with respect to multiple parameters on denoising and deconvolution experiments.",

author = "Tan, {Hong Ye} and Subhadip Mukherjee and Junqi Tang and Andreas Hauptmann and Carola-Bibiane Sch{\"o}nlieb",

year = "2023",

month = may,

day = "5",

doi = "10.1109/ICASSP49357.2023.10096875",

language = "English",

isbn = "9781728163284",

booktitle = "ICASSP 2023 - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",

publisher = "IEEE",

note = "2023 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 ; Conference date: 04-06-2023 Through 10-06-2023",

}

TY - GEN

T1 - Robust Data-Driven Accelerated Mirror Descent

AU - Tan, Hong Ye

AU - Mukherjee, Subhadip

AU - Tang, Junqi

AU - Hauptmann, Andreas

AU - Schönlieb, Carola-Bibiane

PY - 2023/5/5

Y1 - 2023/5/5

N2 - Learning-to-optimize is an emerging framework that leverages training data to speed up the solution of certain optimization problems. One such approach is based on the classical mirror descent algorithm, where the mirror map is modelled using input-convex neural networks. In this work, we extend this functional parameterization approach by introducing momentum into the iterations, based on the classical accelerated mirror descent. Our approach combines short-time accelerated convergence with stable long-time behavior. We empirically demonstrate additional robustness with respect to multiple parameters on denoising and deconvolution experiments.

AB - Learning-to-optimize is an emerging framework that leverages training data to speed up the solution of certain optimization problems. One such approach is based on the classical mirror descent algorithm, where the mirror map is modelled using input-convex neural networks. In this work, we extend this functional parameterization approach by introducing momentum into the iterations, based on the classical accelerated mirror descent. Our approach combines short-time accelerated convergence with stable long-time behavior. We empirically demonstrate additional robustness with respect to multiple parameters on denoising and deconvolution experiments.

U2 - 10.1109/ICASSP49357.2023.10096875

DO - 10.1109/ICASSP49357.2023.10096875

M3 - Conference contribution

SN - 9781728163284

BT - ICASSP 2023 - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

PB - IEEE

T2 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 4 June 2023 through 10 June 2023

ER -

Robust Data-Driven Accelerated Mirror Descent

Abstract

Conference

Access to Document

Fingerprint

Cite this