Discrete-time dynamic average consensus

By Minghui Zhu, Sonia Martínez. at - Automatisierungstechnik, 2010

476 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/736717111417b0637875da42a619bd5d8aa21fd6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

1. Angelia Nedić
2. Alexander Olshevsky
3. Wei Shi
4. Ali H. Sayed
5. Ming Yan
6. P. Lorenzo
7. Zhi Li
8. W. Shi

In Control and Systems Engineering

1. Karl Henrik Johansson
2. Sandro Zampieri
3. Ziyang Meng
4. Xinlei Yi
5. Zongli Lin
6. Wei Shi
7. Tao Yang
8. Ruggero Carli

Citations in the corpus

• Achieving Geometric Convergence for Distributed Optimization Over Time-Varying Graphs (2016, methodology): “The proposed algorithm is based on the combination of Adapt-then-Combine (ATC) strategy of [59] and the dynamic average consensus for gradient tracking of [77].”
• Optimal Synchronization for Networks of Noisy Double Integrators (2011, methodology): “Work in the context of signal tracking with consensus algorithms have been proposed, among others, by Spanos et al. [12], Freeman et al. [3] and Zhu et al. [ 16 ].”
• A Decentralized Proximal-Gradient Method With Network Independent Step-Sizes and Separated Convergence Rates (2017, background): “A number of recent works employed the so-called gradient tracking [37] to conquer different issues [38]–[42].”
• A Framework for Parallel and Distributed Training of Neural Networks (2016, methodology): “Second, the framework exploits a dynamic consensus procedure (Zhu and Mart́ınez, 2010), so that each agent can recover the information relative to all the other agents, which typically is not available at its local side.”
• A survey of distributed optimization (2019, methodology): “More specifically, the update (6a) is a distributed inexact gradient method, where the variable y i ( k ) is used instead of the average gradient, and in the update (6b) , y i ( k ) tracks the average gradient by employing dynamic average consensus ( Zhu & Martínez, 2010 ).”
• Augmented distributed gradient methods for multi-agent optimization under uncoordinated constant stepsizes (2015, unknown): “Lemma 1 (Conservation Property [30]): Let ḡ(k) = 11 m g(x(k)), y(0) = g(x(0)), then ȳ(k) = ḡ(k),∀k ≥ 0.”