Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286977191> ?p ?o ?g. }
Showing items 1 to 66 of
66
with 100 items per page.
- W4286977191 abstract "We consider increasingly complex models of matrix denoising and dictionary learning in the Bayes-optimal setting, in the challenging regime where the matrices to infer have a rank growing linearly with the system size. This is in contrast with most existing literature concerned with the low-rank (i.e., constant-rank) regime. We first consider a class of rotationally invariant matrix denoising problems whose mutual information and minimum mean-square error are computable using techniques from random matrix theory. Next, we analyze the more challenging models of dictionary learning. To do so we introduce a novel combination of the replica method from statistical mechanics together with random matrix theory, coined spectral replica method. This allows us to derive variational formulas for the mutual information between hidden representations and the noisy data of the dictionary learning problem, as well as for the overlaps quantifying the optimal reconstruction error. The proposed method reduces the number of degrees of freedom from $Theta(N^2)$ matrix entries to $Theta(N)$ eigenvalues (or singular values), and yields Coulomb gas representations of the mutual information which are reminiscent of matrix models in physics. The main ingredients are a combination of large deviation results for random matrices together with a new replica symmetric decoupling ansatz at the level of the probability distributions of eigenvalues (or singular values) of certain overlap matrices and the use of HarishChandra-Itzykson-Zuber spherical integrals." @default.
- W4286977191 created "2022-07-25" @default.
- W4286977191 creator A5038854134 @default.
- W4286977191 creator A5057057294 @default.
- W4286977191 date "2021-09-14" @default.
- W4286977191 modified "2023-09-24" @default.
- W4286977191 title "Statistical limits of dictionary learning: random matrix theory and the spectral replica method" @default.
- W4286977191 doi "https://doi.org/10.48550/arxiv.2109.06610" @default.
- W4286977191 hasPublicationYear "2021" @default.
- W4286977191 type Work @default.
- W4286977191 citedByCount "0" @default.
- W4286977191 crossrefType "posted-content" @default.
- W4286977191 hasAuthorship W4286977191A5038854134 @default.
- W4286977191 hasAuthorship W4286977191A5057057294 @default.
- W4286977191 hasConcept C106487976 @default.
- W4286977191 hasConcept C121332964 @default.
- W4286977191 hasConcept C130979935 @default.
- W4286977191 hasConcept C142362112 @default.
- W4286977191 hasConcept C153349607 @default.
- W4286977191 hasConcept C155281189 @default.
- W4286977191 hasConcept C158693339 @default.
- W4286977191 hasConcept C159985019 @default.
- W4286977191 hasConcept C190470478 @default.
- W4286977191 hasConcept C192562407 @default.
- W4286977191 hasConcept C202444582 @default.
- W4286977191 hasConcept C2775937380 @default.
- W4286977191 hasConcept C28826006 @default.
- W4286977191 hasConcept C33923547 @default.
- W4286977191 hasConcept C37914503 @default.
- W4286977191 hasConcept C62520636 @default.
- W4286977191 hasConcept C64812099 @default.
- W4286977191 hasConcept C90199385 @default.
- W4286977191 hasConceptScore W4286977191C106487976 @default.
- W4286977191 hasConceptScore W4286977191C121332964 @default.
- W4286977191 hasConceptScore W4286977191C130979935 @default.
- W4286977191 hasConceptScore W4286977191C142362112 @default.
- W4286977191 hasConceptScore W4286977191C153349607 @default.
- W4286977191 hasConceptScore W4286977191C155281189 @default.
- W4286977191 hasConceptScore W4286977191C158693339 @default.
- W4286977191 hasConceptScore W4286977191C159985019 @default.
- W4286977191 hasConceptScore W4286977191C190470478 @default.
- W4286977191 hasConceptScore W4286977191C192562407 @default.
- W4286977191 hasConceptScore W4286977191C202444582 @default.
- W4286977191 hasConceptScore W4286977191C2775937380 @default.
- W4286977191 hasConceptScore W4286977191C28826006 @default.
- W4286977191 hasConceptScore W4286977191C33923547 @default.
- W4286977191 hasConceptScore W4286977191C37914503 @default.
- W4286977191 hasConceptScore W4286977191C62520636 @default.
- W4286977191 hasConceptScore W4286977191C64812099 @default.
- W4286977191 hasConceptScore W4286977191C90199385 @default.
- W4286977191 hasLocation W42869771911 @default.
- W4286977191 hasOpenAccess W4286977191 @default.
- W4286977191 hasPrimaryLocation W42869771911 @default.
- W4286977191 hasRelatedWork W1724331128 @default.
- W4286977191 hasRelatedWork W1985027133 @default.
- W4286977191 hasRelatedWork W2011934951 @default.
- W4286977191 hasRelatedWork W2025511434 @default.
- W4286977191 hasRelatedWork W2162635298 @default.
- W4286977191 hasRelatedWork W2544357824 @default.
- W4286977191 hasRelatedWork W2734975566 @default.
- W4286977191 hasRelatedWork W2736398265 @default.
- W4286977191 hasRelatedWork W2964005990 @default.
- W4286977191 hasRelatedWork W3164506665 @default.
- W4286977191 isParatext "false" @default.
- W4286977191 isRetracted "false" @default.
- W4286977191 workType "article" @default.