cs.GT, cs.SY, eess.SY, math.OC, stat.ML

Last-Iterate Guarantees for Learning in Co-coercive Games

arXiv:2604.19065v1 Announce Type: cross
Abstract: We establish finite-time last-iterate guarantees for vanilla stochastic gradient descent in co-coercive games under noisy feedback. This is a broad class of games that is more general than strongly mon…