Timezone: »

Communication-Efficient Distributed Learning via Lazily Aggregated Quantized Gradients
Jun Sun · Tianyi Chen · Georgios Giannakis · Zaiyue Yang

Tue Dec 10 05:30 PM -- 07:30 PM (PST) @ East Exhibition Hall B + C #106

The present paper develops a novel aggregated gradient approach for distributed machine learning that adaptively compresses the gradient communication. The key idea is to first quantize the computed gradients, and then skip less informative quantized gradient communications by reusing outdated gradients. Quantizing and skipping result in 'lazy' worker-server communications, which justifies the term Lazily Aggregated Quantized gradient that is henceforth abbreviated as LAQ. Our LAQ can provably attain the same linear convergence rate as the gradient descent in the strongly convex case, while effecting major savings in the communication overhead both in transmitted bits as well as in communication rounds. Empirically, experiments with real data corroborate a significant communication reduction compared to existing gradient- and stochastic gradient-based algorithms.

Author Information

Jun Sun (Zhejiang University)
Tianyi Chen (Rensselaer Polytechnic Institute)
Georgios Giannakis (University of Minnesota)
Zaiyue Yang (Southern University of Science and Technology)

More from the Same Authors