Adaptive projected subgradient method and its acceleration techniques

Isao Yamada, Nobuhiko Ogura, Masahiro Yukawa

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

In this paper, we introduce the adaptive projected subgradient method (APSM) that can minimize asymptotically certain sequence of nonnegative unsmooth convex functions over a closed convex set in a real Hilbert space. The main theorem of the method unifies a wide range of set theoretic adaptive filtering schemes, e.g., N L M S, Projected N L M S, Constrained N L M S, APA and Adaptive parallel outer projection algorithm etc., for possibly nonstationary random processes. We also briefly present a pair of acceleration techniques that can improve efficiently the performance of adaptive parallel outer projection algorithm. These include the ideas using optimal supporting hyperplane of the stochastic property set as well as Pairwise Optimal WEight Realization (POWER) of parallel projection. The numerical examples show that the proposed adaptive filtering scheme realizes dramatically fast and stable convergence for highly coloured excited speech like input signals in severely noisy situations.

Original languageEnglish
Pages (from-to)639-644
Number of pages6
JournalIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume37
Issue number12
DOIs
Publication statusPublished - 2004 Jan 1
Externally publishedYes
Event2004 IFAC Workshop on Adaptation and Learning in Control and Signal Processing, ALCOSP 2004 and IFAC Workshop on Periodic Control Systems, PSYCO 2004 - Yokohama, Japan
Duration: 2004 Aug 302004 Sep 1

Fingerprint

Adaptive filtering
Hilbert spaces
Random processes

Keywords

  • Adaptive projected subgradient method (APSM)
  • Adaptive signal processing
  • Asymptotic minimization of sequence of convex functions
  • Optimal supporting hyperplane
  • Pairwise optimal weight realization (POWER).

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Adaptive projected subgradient method and its acceleration techniques. / Yamada, Isao; Ogura, Nobuhiko; Yukawa, Masahiro.

In: IFAC Proceedings Volumes (IFAC-PapersOnline), Vol. 37, No. 12, 01.01.2004, p. 639-644.

Research output: Contribution to journalConference article

@article{f688efd2ceed43e79dbf623f571e7af8,
title = "Adaptive projected subgradient method and its acceleration techniques",
abstract = "In this paper, we introduce the adaptive projected subgradient method (APSM) that can minimize asymptotically certain sequence of nonnegative unsmooth convex functions over a closed convex set in a real Hilbert space. The main theorem of the method unifies a wide range of set theoretic adaptive filtering schemes, e.g., N L M S, Projected N L M S, Constrained N L M S, APA and Adaptive parallel outer projection algorithm etc., for possibly nonstationary random processes. We also briefly present a pair of acceleration techniques that can improve efficiently the performance of adaptive parallel outer projection algorithm. These include the ideas using optimal supporting hyperplane of the stochastic property set as well as Pairwise Optimal WEight Realization (POWER) of parallel projection. The numerical examples show that the proposed adaptive filtering scheme realizes dramatically fast and stable convergence for highly coloured excited speech like input signals in severely noisy situations.",
keywords = "Adaptive projected subgradient method (APSM), Adaptive signal processing, Asymptotic minimization of sequence of convex functions, Optimal supporting hyperplane, Pairwise optimal weight realization (POWER).",
author = "Isao Yamada and Nobuhiko Ogura and Masahiro Yukawa",
year = "2004",
month = "1",
day = "1",
doi = "10.1016/S1474-6670(17)31541-0",
language = "English",
volume = "37",
pages = "639--644",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "12",

}

TY - JOUR

T1 - Adaptive projected subgradient method and its acceleration techniques

AU - Yamada, Isao

AU - Ogura, Nobuhiko

AU - Yukawa, Masahiro

PY - 2004/1/1

Y1 - 2004/1/1

N2 - In this paper, we introduce the adaptive projected subgradient method (APSM) that can minimize asymptotically certain sequence of nonnegative unsmooth convex functions over a closed convex set in a real Hilbert space. The main theorem of the method unifies a wide range of set theoretic adaptive filtering schemes, e.g., N L M S, Projected N L M S, Constrained N L M S, APA and Adaptive parallel outer projection algorithm etc., for possibly nonstationary random processes. We also briefly present a pair of acceleration techniques that can improve efficiently the performance of adaptive parallel outer projection algorithm. These include the ideas using optimal supporting hyperplane of the stochastic property set as well as Pairwise Optimal WEight Realization (POWER) of parallel projection. The numerical examples show that the proposed adaptive filtering scheme realizes dramatically fast and stable convergence for highly coloured excited speech like input signals in severely noisy situations.

AB - In this paper, we introduce the adaptive projected subgradient method (APSM) that can minimize asymptotically certain sequence of nonnegative unsmooth convex functions over a closed convex set in a real Hilbert space. The main theorem of the method unifies a wide range of set theoretic adaptive filtering schemes, e.g., N L M S, Projected N L M S, Constrained N L M S, APA and Adaptive parallel outer projection algorithm etc., for possibly nonstationary random processes. We also briefly present a pair of acceleration techniques that can improve efficiently the performance of adaptive parallel outer projection algorithm. These include the ideas using optimal supporting hyperplane of the stochastic property set as well as Pairwise Optimal WEight Realization (POWER) of parallel projection. The numerical examples show that the proposed adaptive filtering scheme realizes dramatically fast and stable convergence for highly coloured excited speech like input signals in severely noisy situations.

KW - Adaptive projected subgradient method (APSM)

KW - Adaptive signal processing

KW - Asymptotic minimization of sequence of convex functions

KW - Optimal supporting hyperplane

KW - Pairwise optimal weight realization (POWER).

UR - http://www.scopus.com/inward/record.url?scp=33947106630&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947106630&partnerID=8YFLogxK

U2 - 10.1016/S1474-6670(17)31541-0

DO - 10.1016/S1474-6670(17)31541-0

M3 - Conference article

VL - 37

SP - 639

EP - 644

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 12

ER -