MindFlayer SGD: Efficient Parallel SGD in the Presence of Heterogeneous and Random Worker Compute TimesConference on Uncertainty in Artificial Intelligence (UAI), 2024 |
Fast and Straggler-Tolerant Distributed SGD with Reduced Computation
LoadInternational Symposium on Information Theory (ISIT), 2023 |
Straggler Mitigation through Unequal Error Protection for Distributed
Approximate Matrix MultiplicationIEEE Journal on Selected Areas in Communications (JSAC), 2021 |
On Gradient Coding with Partial RecoveryInternational Symposium on Information Theory (ISIT), 2021 |