Zeroth-Order Diffusion Adaptive Filter Over Networks
- 31 December 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 69 (1053587X) , 589-602
Abstract
Most existing diffusion-based estimation algorithms require the explicit expression of the cost function in order to evaluate the stochastic gradient. In this work, we first discuss the zeroth-order (ZO) gradient for diffusion strategies, and present the ZO-diffusion algorithm that is suitable for applications where the explicit expression of the cost function is unavailable. In addition, to improve the convergence rate of the ZO-diffusion algorithm, we introduce a time-averaging stochastic variance reduced gradient (TA-SVRG) strategy, which is a variant of SVRG algorithm and designed to address online learning problems, and propose a ZO-TA-SVRG diffusion algorithm. Then, we analyze the mean and mean-square stability of the proposed ZO-TA-SVRG diffusion algorithm. Finally, simulation results are provided that demonstrate the performance and effectiveness of the proposed algorithms.Keywords
Funding Information
- National Key Research and Development Program of China Stem Cell and Translational Research (2018AAA0102200)
- National Natural Science Foundation of China (61671382, 61811530283, 61901318)
- Higher Education Discipline Innovation Project (B18041)
- Natural Science Foundation of Jiangsu Province (BK20191419)
- Northwestern Polytechnical University
This publication has 36 references indexed in Scilit:
- Random Gradient-Free Minimization of Convex FunctionsFoundations of Computational Mathematics, 2015
- Diffusion LMS Over Multitask NetworksIEEE Transactions on Signal Processing, 2015
- Optimal Rates for Zero-Order Convex Optimization: The Power of Two Function EvaluationsIEEE Transactions on Information Theory, 2015
- Efficient mini-batch training for stochastic optimization2014
- Multitask Diffusion Adaptation Over NetworksIEEE Transactions on Signal Processing, 2014
- Adaptive NetworksProceedings of the IEEE, 2014
- Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic ProgrammingSIAM Journal on Optimization, 2013
- Derivative-free optimization: a review of algorithms and comparison of software implementationsJournal of Global Optimization, 2012
- Incremental Subgradient Methods for Nondifferentiable OptimizationSIAM Journal on Optimization, 2001
- Mersenne twisterACM Transactions on Modeling and Computer Simulation, 1998