Decentralized Dictionary Learning Over Time-Varying Digraphs

被引:0
作者
Daneshmand, Amir [1 ]
Sun, Ying [1 ]
Scutari, Gesualdo [1 ]
Facchinei, Francisco [2 ]
Sadler, Brian M. [3 ]
机构
[1] Purdue Univ, Sch Ind Engn, W Lafayette, IN 47907 USA
[2] Univ Roma La Sapienza, Dept Comp Control & Management Engn, Rome, Italy
[3] US Army, Res Lab, Adelphi, MD USA
基金
美国国家科学基金会;
关键词
Decentralized algorithms; dictionary learning; directed graph; non-convex optimization; time-varying network; MATRIX FACTORIZATION; SPARSE; ALGORITHM; OPTIMIZATION; CONSENSUS; CONVEX;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper studies Dictionary Learning problems wherein the learning task is distributed over a multi-agent network, modeled as a time-varying directed graph. This formulation is relevant, for instance, in Big Data scenarios where massive amounts of data are collected/stored in different locations (e.g., sensors, clouds) and aggregating and/or processing all data in a fusion center might be inefficient or unfeasible, due to resource limitations, communication overheads or privacy issues. We develop a unified decentralized algorithmic framework for this class of nonconvex problems, which is proved to converge to stationary solutions at a sublinear rate. The new method hinges on Successive Convex Approximation techniques, coupled with a decentralized tracking mechanism aiming at locally estimating the gradient of the smooth part of the sum-utility. To the best of our knowledge, this is the first provably convergent decentralized algorithm for Dictionary Learning and, more generally, bi-convex problems over (time-varying) (di)graphs.
引用
收藏
页数:62
相关论文
共 50 条
[1]   K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J].
Aharon, Michal ;
Elad, Michael ;
Bruckstein, Alfred .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (11) :4311-4322
[2]  
[Anonymous], MISC IM DAT
[3]  
[Anonymous], NIPS
[4]  
[Anonymous], COMPUT SCI REV
[5]  
[Anonymous], MATH PROGRAMMING
[6]  
[Anonymous], 2008, THESIS
[7]  
[Anonymous], THESIS
[8]  
[Anonymous], 2008, P ADV NEURAL INFORM
[9]  
Aravkin A, 2014, UNCERTAINTY IN ARTIFICIAL INTELLIGENCE, P32
[10]  
Bertsekas D., 2015, Parallel and Distributed Computation: Numerical Methods