Matches in SemOpenAlex for { <https://semopenalex.org/work/W3090392341> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W3090392341 abstract "Pre-trained convolutional neural networks (CNNs) are powerful off-the-shelf feature generators and have been shown to perform very well on a variety of tasks. Unfortunately, the generated features are high dimensional and expensive to store: potentially hundreds of thousands of floats per example when processing videos. Traditional entropy based lossless compression methods are of little help as they do not yield desired level of compression, while general purpose lossy compression methods based on energy compaction (e.g. PCA followed by quantization and entropy coding) are sub-optimal, as they are not tuned to task specific objective. We propose a learned method that jointly optimizes for compressibility along with the task objective for learning the features. The plug-in nature of our method makes it straight-forward to integrate with any target objective and trade-off against compressibility. We present results on multiple benchmarks and demonstrate that our method produces features that are an order of magnitude more compressible, while having a regularization effect that leads to a consistent improvement in accuracy." @default.
- W3090392341 created "2020-10-08" @default.
- W3090392341 creator A5024043349 @default.
- W3090392341 creator A5026518815 @default.
- W3090392341 creator A5031745411 @default.
- W3090392341 creator A5069379894 @default.
- W3090392341 creator A5075343029 @default.
- W3090392341 creator A5078347851 @default.
- W3090392341 date "2020-10-01" @default.
- W3090392341 modified "2023-10-16" @default.
- W3090392341 title "End-to-End Learning of Compressible Features" @default.
- W3090392341 cites W1939575207 @default.
- W3090392341 cites W1995875735 @default.
- W3090392341 cites W2034231890 @default.
- W3090392341 cites W2050712059 @default.
- W3090392341 cites W2062118960 @default.
- W3090392341 cites W2095242101 @default.
- W3090392341 cites W2100495367 @default.
- W3090392341 cites W2102605133 @default.
- W3090392341 cites W2108598243 @default.
- W3090392341 cites W2119047110 @default.
- W3090392341 cites W2124509324 @default.
- W3090392341 cites W2133728753 @default.
- W3090392341 cites W2156360858 @default.
- W3090392341 cites W2160921898 @default.
- W3090392341 cites W2464915613 @default.
- W3090392341 cites W2490784316 @default.
- W3090392341 cites W2578797046 @default.
- W3090392341 doi "https://doi.org/10.1109/icip40778.2020.9190860" @default.
- W3090392341 hasPublicationYear "2020" @default.
- W3090392341 type Work @default.
- W3090392341 sameAs 3090392341 @default.
- W3090392341 citedByCount "19" @default.
- W3090392341 countsByYear W30903923412020 @default.
- W3090392341 countsByYear W30903923412021 @default.
- W3090392341 countsByYear W30903923412022 @default.
- W3090392341 countsByYear W30903923412023 @default.
- W3090392341 crossrefType "proceedings-article" @default.
- W3090392341 hasAuthorship W3090392341A5024043349 @default.
- W3090392341 hasAuthorship W3090392341A5026518815 @default.
- W3090392341 hasAuthorship W3090392341A5031745411 @default.
- W3090392341 hasAuthorship W3090392341A5069379894 @default.
- W3090392341 hasAuthorship W3090392341A5075343029 @default.
- W3090392341 hasAuthorship W3090392341A5078347851 @default.
- W3090392341 hasBestOaLocation W30903923412 @default.
- W3090392341 hasConcept C106301342 @default.
- W3090392341 hasConcept C119857082 @default.
- W3090392341 hasConcept C121332964 @default.
- W3090392341 hasConcept C127413603 @default.
- W3090392341 hasConcept C146978453 @default.
- W3090392341 hasConcept C153180895 @default.
- W3090392341 hasConcept C154945302 @default.
- W3090392341 hasConcept C165021410 @default.
- W3090392341 hasConcept C1769480 @default.
- W3090392341 hasConcept C41008148 @default.
- W3090392341 hasConcept C62520636 @default.
- W3090392341 hasConcept C78548338 @default.
- W3090392341 hasConcept C81081738 @default.
- W3090392341 hasConcept C81363708 @default.
- W3090392341 hasConcept C84655787 @default.
- W3090392341 hasConceptScore W3090392341C106301342 @default.
- W3090392341 hasConceptScore W3090392341C119857082 @default.
- W3090392341 hasConceptScore W3090392341C121332964 @default.
- W3090392341 hasConceptScore W3090392341C127413603 @default.
- W3090392341 hasConceptScore W3090392341C146978453 @default.
- W3090392341 hasConceptScore W3090392341C153180895 @default.
- W3090392341 hasConceptScore W3090392341C154945302 @default.
- W3090392341 hasConceptScore W3090392341C165021410 @default.
- W3090392341 hasConceptScore W3090392341C1769480 @default.
- W3090392341 hasConceptScore W3090392341C41008148 @default.
- W3090392341 hasConceptScore W3090392341C62520636 @default.
- W3090392341 hasConceptScore W3090392341C78548338 @default.
- W3090392341 hasConceptScore W3090392341C81081738 @default.
- W3090392341 hasConceptScore W3090392341C81363708 @default.
- W3090392341 hasConceptScore W3090392341C84655787 @default.
- W3090392341 hasLocation W30903923411 @default.
- W3090392341 hasLocation W30903923412 @default.
- W3090392341 hasOpenAccess W3090392341 @default.
- W3090392341 hasPrimaryLocation W30903923411 @default.
- W3090392341 hasRelatedWork W1489137 @default.
- W3090392341 hasRelatedWork W1503901511 @default.
- W3090392341 hasRelatedWork W1790039671 @default.
- W3090392341 hasRelatedWork W2059519944 @default.
- W3090392341 hasRelatedWork W2126219189 @default.
- W3090392341 hasRelatedWork W2155738200 @default.
- W3090392341 hasRelatedWork W2539164965 @default.
- W3090392341 hasRelatedWork W2556577032 @default.
- W3090392341 hasRelatedWork W2977360967 @default.
- W3090392341 hasRelatedWork W3008260912 @default.
- W3090392341 isParatext "false" @default.
- W3090392341 isRetracted "false" @default.
- W3090392341 magId "3090392341" @default.
- W3090392341 workType "article" @default.