Majorization-Minimization Algorithms in Signal Processing, Communications, and Machine Learning

This paper gives an overview of the majorization-minimization (MM) algorithmic framework, which can provide guidance in deriving problem-driven algorithms with low computational cost. A general introduction of MM is presented, including a description of the basic principle and its convergence result...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on signal processing Ročník 65; číslo 3; s. 794 - 816
Hlavní autoři: Ying Sun, Babu, Prabhu, Palomar, Daniel P.
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.02.2017
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algorithms
Basic converters
Convergence
Convexity
Estimation
Linear programming
Machine learning
Majorization-minimization
Minimization
non-convex optimization
Optimization
Signal processing
Signal processing algorithms
surrogate function
Taylor series
Upper bounds
upperbounds
ISSN:1053-587X, 1941-0476
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper gives an overview of the majorization-minimization (MM) algorithmic framework, which can provide guidance in deriving problem-driven algorithms with low computational cost. A general introduction of MM is presented, including a description of the basic principle and its convergence results. The extensions, acceleration schemes, and connection to other algorithmic frameworks are also covered. To bridge the gap between theory and practice, upperbounds for a large number of basic functions, derived based on the Taylor expansion, convexity, and special inequalities, are provided as ingredients for constructing surrogate functions. With the pre-requisites established, the way of applying MM to solving specific problems is elaborated by a wide range of applications in signal processing, communications, and machine learning.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1053-587X
1941-0476
DOI:10.1109/TSP.2016.2601299