{"id":"arxiv/2605.28980","url":"https://arxiv.org/abs/2605.28980","tags":["math.OC","cs.LG","eess.SP","math.NA"],"score":0,"title":"Manifold-based Algorithms for the Hadamard Decomposition","domain":"math.OC","source":"arxiv","authors":["Nicolas Gillis","Subhayan Saha","Stefano Sicilia","Arnaud Vandaele"],"reviews":0,"subject":"math.OC","summary":"Given a matrix $X$, and two ranks $r_1$ and $r_2$, the Hadamard decomposition (HD) looks for two low-rank matrices, $X_1$ of rank $r_1$ and $X_2$ of rank $r_2$, both of the same size as $X$, such that $X\\approx X_1\\circ X_2$, where $\\circ$ is the Hadamard (element-wise) product. In most cases, HD is more expressive than standard low-rank approximations such as the truncated singular value decomposition (TSVD), as it can represent higher-rank matrices with the same number of parameters; this is because the rank of $X_1 \\circ X_2$ is generically equal to $r_1 r_2$. In this paper, we first…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-27","reactions":{}}