Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
A Tight Theory of Error Feedback Algorithms in Distributed Optimization
Daniel Berg Thomsen, Adrien Taylor, Aymeric Dieuleveut
Communication costs are a major bottleneck in distributed learning and first-order optimization. A common approach to alleviate this issue is to compress the gradient information e…
cs.LG2025
Tight analyses of first-order methods with error feedback
Daniel Berg Thomsen, Adrien Taylor, Aymeric Dieuleveut
Communication between agents often constitutes a major computational bottleneck in distributed learning. One of the most common mitigation strategies is to compress the information…