Matches in SemOpenAlex for { <https://semopenalex.org/work/W3183420147> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W3183420147 endingPage "49" @default.
- W3183420147 startingPage "33" @default.
- W3183420147 abstract "Recently proposed neural architecture search (NAS) algorithms adopt neural predictors to accelerate architecture search. The capability of neural predictors to accurately predict the performance metrics of the neural architecture is critical to NAS, but obtaining training datasets for neural predictors is often time-consuming. How to obtain a neural predictor with high prediction accuracy using a small amount of training data is a central problem to neural predictor-based NAS. Here, a new architecture encoding scheme is first devised to calculate the graph edit distance of neural architectures, which overcomes the drawbacks of existing vector-based architecture encoding schemes. To enhance the predictive performance of neural predictors, two self-supervised learning methods are proposed to pre-train the architecture embedding part of neural predictors to generate a meaningful representation of neural architectures. The first method designs a graph neural network-based model with two independent branches and utilizes the graph edit distance of two different neural architectures as a supervision to force the model to generate meaningful architecture representations. Inspired by contrastive learning, the second method presents a new contrastive learning algorithm that utilizes a central feature vector as a proxy to contrast positive pairs against negative pairs. Experimental results illustrate that the pre-trained neural predictors can achieve comparable or superior performance compared with their supervised counterparts using only half of the training samples. The effectiveness of the proposed methods is further validated by integrating the pre-trained neural predictors into a neural predictor guided evolutionary neural architecture search (NPENAS) algorithm, which achieves stateof-the-art performance on NASBench-101, NASBench-201, and DARTS benchmarks" @default.
- W3183420147 created "2021-08-02" @default.
- W3183420147 creator A5010189817 @default.
- W3183420147 creator A5032808299 @default.
- W3183420147 creator A5033632697 @default.
- W3183420147 creator A5037855900 @default.
- W3183420147 creator A5065193592 @default.
- W3183420147 creator A5077260423 @default.
- W3183420147 date "2021-08-01" @default.
- W3183420147 modified "2023-09-26" @default.
- W3183420147 title "Self-Supervised Representation Learning for Evolutionary Neural Architecture Search" @default.
- W3183420147 cites W2321533354 @default.
- W3183420147 cites W2887063112 @default.
- W3183420147 cites W2906943923 @default.
- W3183420147 cites W2953308748 @default.
- W3183420147 cites W2963946985 @default.
- W3183420147 cites W2964081807 @default.
- W3183420147 cites W2964544183 @default.
- W3183420147 cites W2965658867 @default.
- W3183420147 cites W2997699039 @default.
- W3183420147 cites W3035524453 @default.
- W3183420147 cites W3094801149 @default.
- W3183420147 cites W3107453328 @default.
- W3183420147 cites W4210257598 @default.
- W3183420147 doi "https://doi.org/10.1109/mci.2021.3084415" @default.
- W3183420147 hasPublicationYear "2021" @default.
- W3183420147 type Work @default.
- W3183420147 sameAs 3183420147 @default.
- W3183420147 citedByCount "7" @default.
- W3183420147 countsByYear W31834201472022 @default.
- W3183420147 countsByYear W31834201472023 @default.
- W3183420147 crossrefType "journal-article" @default.
- W3183420147 hasAuthorship W3183420147A5010189817 @default.
- W3183420147 hasAuthorship W3183420147A5032808299 @default.
- W3183420147 hasAuthorship W3183420147A5033632697 @default.
- W3183420147 hasAuthorship W3183420147A5037855900 @default.
- W3183420147 hasAuthorship W3183420147A5065193592 @default.
- W3183420147 hasAuthorship W3183420147A5077260423 @default.
- W3183420147 hasBestOaLocation W31834201472 @default.
- W3183420147 hasConcept C108583219 @default.
- W3183420147 hasConcept C119857082 @default.
- W3183420147 hasConcept C123657996 @default.
- W3183420147 hasConcept C125411270 @default.
- W3183420147 hasConcept C132525143 @default.
- W3183420147 hasConcept C142362112 @default.
- W3183420147 hasConcept C153180895 @default.
- W3183420147 hasConcept C153349607 @default.
- W3183420147 hasConcept C154945302 @default.
- W3183420147 hasConcept C175202392 @default.
- W3183420147 hasConcept C41008148 @default.
- W3183420147 hasConcept C41608201 @default.
- W3183420147 hasConcept C50644808 @default.
- W3183420147 hasConcept C80444323 @default.
- W3183420147 hasConceptScore W3183420147C108583219 @default.
- W3183420147 hasConceptScore W3183420147C119857082 @default.
- W3183420147 hasConceptScore W3183420147C123657996 @default.
- W3183420147 hasConceptScore W3183420147C125411270 @default.
- W3183420147 hasConceptScore W3183420147C132525143 @default.
- W3183420147 hasConceptScore W3183420147C142362112 @default.
- W3183420147 hasConceptScore W3183420147C153180895 @default.
- W3183420147 hasConceptScore W3183420147C153349607 @default.
- W3183420147 hasConceptScore W3183420147C154945302 @default.
- W3183420147 hasConceptScore W3183420147C175202392 @default.
- W3183420147 hasConceptScore W3183420147C41008148 @default.
- W3183420147 hasConceptScore W3183420147C41608201 @default.
- W3183420147 hasConceptScore W3183420147C50644808 @default.
- W3183420147 hasConceptScore W3183420147C80444323 @default.
- W3183420147 hasIssue "3" @default.
- W3183420147 hasLocation W31834201471 @default.
- W3183420147 hasLocation W31834201472 @default.
- W3183420147 hasOpenAccess W3183420147 @default.
- W3183420147 hasPrimaryLocation W31834201471 @default.
- W3183420147 hasRelatedWork W3014300295 @default.
- W3183420147 hasRelatedWork W3156786002 @default.
- W3183420147 hasRelatedWork W3195829100 @default.
- W3183420147 hasRelatedWork W4223943233 @default.
- W3183420147 hasRelatedWork W4225161397 @default.
- W3183420147 hasRelatedWork W4309045103 @default.
- W3183420147 hasRelatedWork W4312200629 @default.
- W3183420147 hasRelatedWork W4360585206 @default.
- W3183420147 hasRelatedWork W4364306694 @default.
- W3183420147 hasRelatedWork W1629725936 @default.
- W3183420147 hasVolume "16" @default.
- W3183420147 isParatext "false" @default.
- W3183420147 isRetracted "false" @default.
- W3183420147 magId "3183420147" @default.
- W3183420147 workType "article" @default.