Title | ||
---|---|---|
Communication-Efficient and Byzantine-Robust Distributed Learning With Error Feedback |
Abstract | ||
---|---|---|
We develop a communication-efficient distributed learning algorithm that is robust against Byzantine worker machines. We propose and analyze a distributed gradient-descent algorithm that performs a simple thresholding based on gradient norms to mitigate Byzantine failures. We show the (statistical) error-rate of our algorithm matches that of Yin et al. (2018), which uses more comp... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/JSAIT.2021.3105076 | IEEE Journal on Selected Areas in Information Theory |
Keywords | DocType | Volume |
Convergence,Error analysis,Compressors,Manganese,Information theory,Resilience,Neural networks | Journal | 2 |
Issue | Citations | PageRank |
3 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Avishek Ghosh | 1 | 32 | 11.35 |
Raj Kumar Maity | 2 | 0 | 1.69 |
Swanand Kadhe | 3 | 54 | 11.22 |
Arya Mazumdar | 4 | 307 | 41.81 |
Kannan Ramchandran | 5 | 9401 | 1029.57 |