Generalization Error Matters in Decentralized Learning Under Byzantine Attacks
This addresses a critical gap for practitioners implementing resilient decentralized learning systems, though it is incremental as it builds on existing Byzantine-resilient DSGD algorithms.
The paper tackles the problem of analyzing generalization errors in decentralized learning under Byzantine attacks, revealing that these errors persist even with infinite training samples due to malicious agents.
Recently, decentralized learning has emerged as a popular peer-to-peer signal and information processing paradigm that enables model training across geographically distributed agents in a scalable manner, without the presence of any central server. When some of the agents are malicious (also termed as Byzantine), resilient decentralized learning algorithms are able to limit the impact of these Byzantine agents without knowing their number and identities, and have guaranteed optimization errors. However, analysis of the generalization errors, which are critical to implementations of the trained models, is still lacking. In this paper, we provide the first analysis of the generalization errors for a class of popular Byzantine-resilient decentralized stochastic gradient descent (DSGD) algorithms. Our theoretical results reveal that the generalization errors cannot be entirely eliminated because of the presence of the Byzantine agents, even if the number of training samples are infinitely large. Numerical experiments are conducted to confirm our theoretical results.