1. Home
2. Premium
3. IEEE ICASSP 2020 Virtual Conference May 2020
4. Overlap Local-Sgd: An Algorithmic Approach To Hide Communication Delays In Distributed Sgd

Overlap Local-Sgd: An Algorithmic Approach To Hide Communication Delays In Distributed Sgd

Collection:

IEEE ICASSP 2020 Virtual Conference May 2020

Distributed stochastic gradient descent (SGD) is essential for scaling the machine learning algorithms to a large number of computing nodes. However, the infrastructures variability such as high communication delay or random node slowdown greatly impedes

• IEEE MemberUS $11.00
• Society MemberUS $0.00
• IEEE Student MemberUS $11.00
• Non-IEEE MemberUS $15.00

Purchase

Videos in this product

Overlap Local-Sgd: An Algorithmic Approach To Hide Communication Delays In Distributed Sgd

00:15:54

0 views

Distributed stochastic gradient descent (SGD) is essential for scaling the machine learning algorithms to a large number of computing nodes. However, the infrastructures variability such as high communication delay or random node slowdown greatly impedes