Matches in SemOpenAlex for { <https://semopenalex.org/work/W4361285887> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4361285887 endingPage "109561" @default.
- W4361285887 startingPage "109561" @default.
- W4361285887 abstract "The exemplar-free class incremental learning (IL) requires classification models to learn new-class knowledge incrementally without retaining any old samples. Recently, the IL framework based on parallel one-class classifiers (POC) has demonstrated promising performance. It trains a one-class classifier (OCC) for each category and thus is immune to the catastrophic forgetting problem. However, the single-class training strategy may incur weak discriminability and low comparability between different classifiers in POC. To meet this challenge, we propose a new IL framework, referred to as Discriminative and Comparable Parallel One-class Classifiers (DCPOC). Instead of ordinary OCCs (e.g., deep SVDD) used in other POC methods, DCPOC adopts variational auto-encoders (VAE) as OCCs because VAEs can be used not only to identify classes for given samples but also to generate pseudo samples for the trained classes. With this advantage, DCPOC trains a new-class VAE in contrast with the old-class VAEs, which benefits the new-class VAE to reconstruct better for new-class samples but worse for old-class pseudo samples, thus enhancing the comparability. Furthermore, DCPOC introduces a hinge reconstruction loss to reinforce the discriminability. We evaluate our method on MNIST, CIFAR10, CIFAR100, Tiny-ImageNet, and ImageNet. The experimental results show that DCPOC achieves state-of-the-art performance on these datasets.1" @default.
- W4361285887 created "2023-03-31" @default.
- W4361285887 creator A5005350496 @default.
- W4361285887 creator A5017947172 @default.
- W4361285887 creator A5018191388 @default.
- W4361285887 creator A5049242358 @default.
- W4361285887 creator A5075882217 @default.
- W4361285887 creator A5086707829 @default.
- W4361285887 date "2023-08-01" @default.
- W4361285887 modified "2023-09-25" @default.
- W4361285887 title "Exemplar-free class incremental learning via discriminative and comparable parallel one-class classifiers" @default.
- W4361285887 cites W1682403713 @default.
- W4361285887 cites W2100495367 @default.
- W4361285887 cites W2473930607 @default.
- W4361285887 cites W2560647685 @default.
- W4361285887 cites W2967821093 @default.
- W4361285887 cites W3030364939 @default.
- W4361285887 cites W3125116114 @default.
- W4361285887 cites W4225150604 @default.
- W4361285887 doi "https://doi.org/10.1016/j.patcog.2023.109561" @default.
- W4361285887 hasPublicationYear "2023" @default.
- W4361285887 type Work @default.
- W4361285887 citedByCount "2" @default.
- W4361285887 countsByYear W43612858872023 @default.
- W4361285887 crossrefType "journal-article" @default.
- W4361285887 hasAuthorship W4361285887A5005350496 @default.
- W4361285887 hasAuthorship W4361285887A5017947172 @default.
- W4361285887 hasAuthorship W4361285887A5018191388 @default.
- W4361285887 hasAuthorship W4361285887A5049242358 @default.
- W4361285887 hasAuthorship W4361285887A5075882217 @default.
- W4361285887 hasAuthorship W4361285887A5086707829 @default.
- W4361285887 hasConcept C108583219 @default.
- W4361285887 hasConcept C114614502 @default.
- W4361285887 hasConcept C119857082 @default.
- W4361285887 hasConcept C153180895 @default.
- W4361285887 hasConcept C154945302 @default.
- W4361285887 hasConcept C190502265 @default.
- W4361285887 hasConcept C197947376 @default.
- W4361285887 hasConcept C2777212361 @default.
- W4361285887 hasConcept C33923547 @default.
- W4361285887 hasConcept C34872919 @default.
- W4361285887 hasConcept C41008148 @default.
- W4361285887 hasConcept C95623464 @default.
- W4361285887 hasConcept C97931131 @default.
- W4361285887 hasConceptScore W4361285887C108583219 @default.
- W4361285887 hasConceptScore W4361285887C114614502 @default.
- W4361285887 hasConceptScore W4361285887C119857082 @default.
- W4361285887 hasConceptScore W4361285887C153180895 @default.
- W4361285887 hasConceptScore W4361285887C154945302 @default.
- W4361285887 hasConceptScore W4361285887C190502265 @default.
- W4361285887 hasConceptScore W4361285887C197947376 @default.
- W4361285887 hasConceptScore W4361285887C2777212361 @default.
- W4361285887 hasConceptScore W4361285887C33923547 @default.
- W4361285887 hasConceptScore W4361285887C34872919 @default.
- W4361285887 hasConceptScore W4361285887C41008148 @default.
- W4361285887 hasConceptScore W4361285887C95623464 @default.
- W4361285887 hasConceptScore W4361285887C97931131 @default.
- W4361285887 hasFunder F4320321001 @default.
- W4361285887 hasFunder F4320321543 @default.
- W4361285887 hasFunder F4320328119 @default.
- W4361285887 hasLocation W43612858871 @default.
- W4361285887 hasOpenAccess W4361285887 @default.
- W4361285887 hasPrimaryLocation W43612858871 @default.
- W4361285887 hasRelatedWork W2024160000 @default.
- W4361285887 hasRelatedWork W2061273563 @default.
- W4361285887 hasRelatedWork W2088804257 @default.
- W4361285887 hasRelatedWork W2285052147 @default.
- W4361285887 hasRelatedWork W2556319748 @default.
- W4361285887 hasRelatedWork W2729514902 @default.
- W4361285887 hasRelatedWork W2742991909 @default.
- W4361285887 hasRelatedWork W2773500201 @default.
- W4361285887 hasRelatedWork W2810865670 @default.
- W4361285887 hasRelatedWork W4243326899 @default.
- W4361285887 hasVolume "140" @default.
- W4361285887 isParatext "false" @default.
- W4361285887 isRetracted "false" @default.
- W4361285887 workType "article" @default.