Matches in SemOpenAlex for { <https://semopenalex.org/work/W3175422210> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3175422210 endingPage "4007" @default.
- W3175422210 startingPage "3989" @default.
- W3175422210 abstract "The seminal result of Johnson and Lindenstrauss on random embeddings has been intensively studied in applied and theoretical computer science. Despite that vast body of literature, we still lack of complete understanding of statistical properties of random projections; a particularly intriguing question is: why are the theoretical bounds that far behind the empirically observed performance? Motivated by this question, this work develops Johnson-Lindenstrauss distributions with optimal, data-oblivious, statistical confidence bounds. These bounds are numerically best possible, for any given data dimension, embedding dimension, and distortion tolerance. They improve upon prior works in terms of statistical accuracy, as well as exactly determine the no-go regimes for data-oblivious approaches. Furthermore, the projection matrices are efficiently samplable. The construction relies on orthogonal matrices, and the proof uses certain elegant properties of the unit sphere. In particular, the following techniques introduced in this work are of independent interest: a) a compact expression for the projection distortion in terms of singular eigenvalues of the projection matrix, b) a parametrization linking the unit sphere and the Dirichlet distribution and c) anti-concentration bounds for the Dirichlet distribution. Besides the technical contribution, the paper presents applications and numerical evaluation along with working implementation in Python (shared as a GitHub repository)." @default.
- W3175422210 created "2021-07-05" @default.
- W3175422210 creator A5011993100 @default.
- W3175422210 date "2021-07-07" @default.
- W3175422210 modified "2023-09-23" @default.
- W3175422210 title "Johnson-Lindenstrauss Transforms with Best Confidence" @default.
- W3175422210 hasPublicationYear "2021" @default.
- W3175422210 type Work @default.
- W3175422210 sameAs 3175422210 @default.
- W3175422210 citedByCount "0" @default.
- W3175422210 crossrefType "proceedings-article" @default.
- W3175422210 hasAuthorship W3175422210A5011993100 @default.
- W3175422210 hasConcept C111919701 @default.
- W3175422210 hasConcept C11413529 @default.
- W3175422210 hasConcept C114614502 @default.
- W3175422210 hasConcept C121332964 @default.
- W3175422210 hasConcept C134306372 @default.
- W3175422210 hasConcept C154945302 @default.
- W3175422210 hasConcept C158693339 @default.
- W3175422210 hasConcept C169214877 @default.
- W3175422210 hasConcept C182310444 @default.
- W3175422210 hasConcept C2777036070 @default.
- W3175422210 hasConcept C2780803321 @default.
- W3175422210 hasConcept C28826006 @default.
- W3175422210 hasConcept C33676613 @default.
- W3175422210 hasConcept C33923547 @default.
- W3175422210 hasConcept C41008148 @default.
- W3175422210 hasConcept C41608201 @default.
- W3175422210 hasConcept C519991488 @default.
- W3175422210 hasConcept C57273362 @default.
- W3175422210 hasConcept C57493831 @default.
- W3175422210 hasConcept C62520636 @default.
- W3175422210 hasConcept C64812099 @default.
- W3175422210 hasConcept C80444323 @default.
- W3175422210 hasConceptScore W3175422210C111919701 @default.
- W3175422210 hasConceptScore W3175422210C11413529 @default.
- W3175422210 hasConceptScore W3175422210C114614502 @default.
- W3175422210 hasConceptScore W3175422210C121332964 @default.
- W3175422210 hasConceptScore W3175422210C134306372 @default.
- W3175422210 hasConceptScore W3175422210C154945302 @default.
- W3175422210 hasConceptScore W3175422210C158693339 @default.
- W3175422210 hasConceptScore W3175422210C169214877 @default.
- W3175422210 hasConceptScore W3175422210C182310444 @default.
- W3175422210 hasConceptScore W3175422210C2777036070 @default.
- W3175422210 hasConceptScore W3175422210C2780803321 @default.
- W3175422210 hasConceptScore W3175422210C28826006 @default.
- W3175422210 hasConceptScore W3175422210C33676613 @default.
- W3175422210 hasConceptScore W3175422210C33923547 @default.
- W3175422210 hasConceptScore W3175422210C41008148 @default.
- W3175422210 hasConceptScore W3175422210C41608201 @default.
- W3175422210 hasConceptScore W3175422210C519991488 @default.
- W3175422210 hasConceptScore W3175422210C57273362 @default.
- W3175422210 hasConceptScore W3175422210C57493831 @default.
- W3175422210 hasConceptScore W3175422210C62520636 @default.
- W3175422210 hasConceptScore W3175422210C64812099 @default.
- W3175422210 hasConceptScore W3175422210C80444323 @default.
- W3175422210 hasLocation W31754222101 @default.
- W3175422210 hasOpenAccess W3175422210 @default.
- W3175422210 hasPrimaryLocation W31754222101 @default.
- W3175422210 hasRelatedWork W1556058505 @default.
- W3175422210 hasRelatedWork W1657130172 @default.
- W3175422210 hasRelatedWork W1895639643 @default.
- W3175422210 hasRelatedWork W1958559323 @default.
- W3175422210 hasRelatedWork W1993229276 @default.
- W3175422210 hasRelatedWork W2182347905 @default.
- W3175422210 hasRelatedWork W2417169585 @default.
- W3175422210 hasRelatedWork W2594878591 @default.
- W3175422210 hasRelatedWork W2797520557 @default.
- W3175422210 hasRelatedWork W2799483391 @default.
- W3175422210 hasRelatedWork W2803952844 @default.
- W3175422210 hasRelatedWork W2950663854 @default.
- W3175422210 hasRelatedWork W2952921393 @default.
- W3175422210 hasRelatedWork W2991285061 @default.
- W3175422210 hasRelatedWork W2995799124 @default.
- W3175422210 hasRelatedWork W3003739484 @default.
- W3175422210 hasRelatedWork W3083187399 @default.
- W3175422210 hasRelatedWork W3153900045 @default.
- W3175422210 hasRelatedWork W3156917609 @default.
- W3175422210 hasRelatedWork W3189137649 @default.
- W3175422210 isParatext "false" @default.
- W3175422210 isRetracted "false" @default.
- W3175422210 magId "3175422210" @default.
- W3175422210 workType "article" @default.