AI RESEARCH
Error whitening: Why Gauss-Newton outperforms Newton
arXiv CS.LG
•
ArXi:2605.11316v1 Announce Type: new The Gauss-Newton matrix is widely viewed as a positive semidefinite approximation of the Hessian, yet mounting empirical evidence shows that Gauss-Newton descent outperforms Newton's method. We adopt a function space perspective to analyze this phenomenon.