{"id":"arxiv/2602.09240","url":"https://arxiv.org/abs/2602.09240","tags":["math.ST","cs.IT","cs.LG","math.PR","stat.ML"],"score":0,"title":"Optimal Estimation in Orthogonally Invariant Generalized Linear Models: Spectral Initialization and Approximate Message Passing","domain":"math.ST","source":"arxiv","authors":["Yihan Zhang","Hong Chang Ji","Ramji Venkataramanan","Marco Mondelli"],"reviews":0,"subject":"math.ST","summary":"We consider the problem of parameter estimation from a generalized linear model with a random design matrix that is orthogonally invariant in law. Such a model allows the design have an arbitrary distribution of singular values and only assumes that its singular vectors are generic. It is a vast generalization of the i.i.d. Gaussian design typically considered in the theoretical literature, and is motivated by the fact that real data often have a complex correlation structure so that methods relying on i.i.d. assumptions can be highly suboptimal. Building on the paradigm of…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-09","reactions":{}}