Showing 2025Show all
2 papers · 1 filter
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…
math.OC2025
Open Problem: Two Riddles in Heavy-Ball Dynamics
Baptiste Goujaud, Adrien Taylor, Aymeric Dieuleveut
This short paper presents two open problems on the widely used Polyak's Heavy-Ball algorithm. The first problem is the method's ability to exactly \textit{accelerate} in dimension…