Demystifying parallel and distributed deep learning: An in-depth concurrency analysis

T Ben-Nun, T Hoefler - ACM Computing Surveys (CSUR), 2019 - dl.acm.org
Deep Neural Networks (DNNs) are becoming an important tool in modern computing
applications. Accelerating their training is a major challenge and techniques range from …

Collective motion

T Vicsek, A Zafeiris - Physics reports, 2012 - Elsevier
We review the observations and the basic laws describing the essential aspects of collective
motion—being one of the most common and spectacular manifestation of coordinated …

Can decentralized algorithms outperform centralized algorithms? a case study for decentralized parallel stochastic gradient descent

X Lian, C Zhang, H Zhang, CJ Hsieh… - Advances in neural …, 2017 - proceedings.neurips.cc
Most distributed machine learning systems nowadays, including TensorFlow and CNTK, are
built in a centralized fashion. One bottleneck of centralized algorithms lies on high …

Consensus and cooperation in networked multi-agent systems

R Olfati-Saber, JA Fax, RM Murray - Proceedings of the IEEE, 2007 - ieeexplore.ieee.org
This paper provides a theoretical framework for analysis of consensus algorithms for multi-
agent networked systems with an emphasis on the role of directed information flow …

Distributed subgradient methods for multi-agent optimization

A Nedic, A Ozdaglar - IEEE Transactions on Automatic Control, 2009 - ieeexplore.ieee.org
We study a distributed computation model for optimizing a sum of convex objective functions
corresponding to multiple agents. For solving this (not necessarily smooth) optimization …

Randomized gossip algorithms

S Boyd, A Ghosh, B Prabhakar… - IEEE transactions on …, 2006 - ieeexplore.ieee.org
Motivated by applications to sensor, peer-to-peer, and ad hoc networks, we study distributed
algorithms, also known as gossip algorithms, for exchanging information and for computing …

Network topology and communication-computation tradeoffs in decentralized optimization

A Nedić, A Olshevsky, MG Rabbat - Proceedings of the IEEE, 2018 - ieeexplore.ieee.org
In decentralized optimization, nodes cooperate to minimize an overall objective function that
is the sum (or average) of per-node private objective functions. Algorithms interleave local …

Constrained consensus and optimization in multi-agent networks

A Nedic, A Ozdaglar, PA Parrilo - IEEE Transactions on …, 2010 - ieeexplore.ieee.org
We present distributed algorithms that can be used by multiple agents to align their
estimates with a particular value over a network with time-varying connectivity. Our …

Decentralized federated averaging

T Sun, D Li, B Wang - IEEE Transactions on Pattern Analysis …, 2022 - ieeexplore.ieee.org
Federated averaging (FedAvg) is a communication-efficient algorithm for distributed training
with an enormous number of clients. In FedAvg, clients keep their data locally for privacy …

Asynchronous decentralized parallel stochastic gradient descent

X Lian, W Zhang, C Zhang, J Liu - … Conference on Machine …, 2018 - proceedings.mlr.press
Most commonly used distributed machine learning systems are either synchronous or
centralized asynchronous. Synchronous algorithms like AllReduce-SGD perform poorly in a …