Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953783708> ?p ?o ?g. }
- W2953783708 abstract "The ability to learn more and more concepts over time from incrementally arriving data is essential for the development of a life-long learning system. However, deep neural networks often suffer from forgetting previously learned concepts when continually learning new concepts, which is known as catastrophic forgetting problem. The main reason for catastrophic forgetting is that the past concept data is not available and neural weights are changed during incrementally learning new concepts. In this paper, we propose a pseudo-rehearsal based class incremental learning approach to make neural networks capable of continually learning new concepts. We use a conditional generative adversarial network to consolidate old concepts memory and recall pseudo samples during learning new concepts and a balanced online memory recall strategy is to maximally maintain old memories. And we design a comprehensible incremental concept learning network as well as a concept contrastive loss to alleviate the magnitude of neural weights change. We evaluate the proposed approach on MNIST, Fashion-MNIST and SVHN datasets and compare with other rehearsal based approaches. The extensive experiments demonstrate the effectiveness of our approach." @default.
- W2953783708 created "2019-07-12" @default.
- W2953783708 creator A5019329695 @default.
- W2953783708 creator A5019915306 @default.
- W2953783708 creator A5069298091 @default.
- W2953783708 date "2019-07-05" @default.
- W2953783708 modified "2023-09-27" @default.
- W2953783708 title "Incremental Concept Learning via Online Generative Memory Recall" @default.
- W2953783708 cites W1522301498 @default.
- W2953783708 cites W1681844354 @default.
- W2953783708 cites W1682403713 @default.
- W2953783708 cites W1793121960 @default.
- W2953783708 cites W1821462560 @default.
- W2953783708 cites W1979631293 @default.
- W2953783708 cites W2037979274 @default.
- W2953783708 cites W2047057213 @default.
- W2953783708 cites W2099471712 @default.
- W2953783708 cites W2116522068 @default.
- W2953783708 cites W2335728318 @default.
- W2953783708 cites W2473930607 @default.
- W2953783708 cites W2474280151 @default.
- W2953783708 cites W2520774990 @default.
- W2953783708 cites W2523246573 @default.
- W2953783708 cites W2560258536 @default.
- W2953783708 cites W2560647685 @default.
- W2953783708 cites W2593414223 @default.
- W2953783708 cites W2594833348 @default.
- W2953783708 cites W2597948870 @default.
- W2953783708 cites W2605043629 @default.
- W2953783708 cites W2744654301 @default.
- W2953783708 cites W2750384547 @default.
- W2953783708 cites W2765895026 @default.
- W2953783708 cites W2785678896 @default.
- W2953783708 cites W2786498526 @default.
- W2953783708 cites W2788388592 @default.
- W2953783708 cites W2850509156 @default.
- W2953783708 cites W2892234010 @default.
- W2953783708 cites W2893749619 @default.
- W2953783708 cites W2894094671 @default.
- W2953783708 cites W2899771611 @default.
- W2953783708 cites W2921184280 @default.
- W2953783708 cites W2950527759 @default.
- W2953783708 cites W2950893734 @default.
- W2953783708 cites W2951004968 @default.
- W2953783708 cites W2962724315 @default.
- W2953783708 cites W2962879692 @default.
- W2953783708 cites W2963041308 @default.
- W2953783708 cites W2963559848 @default.
- W2953783708 cites W2963788399 @default.
- W2953783708 cites W2964189064 @default.
- W2953783708 cites W2426267443 @default.
- W2953783708 doi "https://doi.org/10.48550/arxiv.1907.02788" @default.
- W2953783708 hasPublicationYear "2019" @default.
- W2953783708 type Work @default.
- W2953783708 sameAs 2953783708 @default.
- W2953783708 citedByCount "0" @default.
- W2953783708 crossrefType "posted-content" @default.
- W2953783708 hasAuthorship W2953783708A5019329695 @default.
- W2953783708 hasAuthorship W2953783708A5019915306 @default.
- W2953783708 hasAuthorship W2953783708A5069298091 @default.
- W2953783708 hasBestOaLocation W29537837081 @default.
- W2953783708 hasConcept C100660578 @default.
- W2953783708 hasConcept C108583219 @default.
- W2953783708 hasConcept C119857082 @default.
- W2953783708 hasConcept C154945302 @default.
- W2953783708 hasConcept C15744967 @default.
- W2953783708 hasConcept C180747234 @default.
- W2953783708 hasConcept C190502265 @default.
- W2953783708 hasConcept C2780735816 @default.
- W2953783708 hasConcept C39890363 @default.
- W2953783708 hasConcept C41008148 @default.
- W2953783708 hasConcept C50644808 @default.
- W2953783708 hasConcept C7149132 @default.
- W2953783708 hasConceptScore W2953783708C100660578 @default.
- W2953783708 hasConceptScore W2953783708C108583219 @default.
- W2953783708 hasConceptScore W2953783708C119857082 @default.
- W2953783708 hasConceptScore W2953783708C154945302 @default.
- W2953783708 hasConceptScore W2953783708C15744967 @default.
- W2953783708 hasConceptScore W2953783708C180747234 @default.
- W2953783708 hasConceptScore W2953783708C190502265 @default.
- W2953783708 hasConceptScore W2953783708C2780735816 @default.
- W2953783708 hasConceptScore W2953783708C39890363 @default.
- W2953783708 hasConceptScore W2953783708C41008148 @default.
- W2953783708 hasConceptScore W2953783708C50644808 @default.
- W2953783708 hasConceptScore W2953783708C7149132 @default.
- W2953783708 hasLocation W29537837081 @default.
- W2953783708 hasOpenAccess W2953783708 @default.
- W2953783708 hasPrimaryLocation W29537837081 @default.
- W2953783708 hasRelatedWork W2597787948 @default.
- W2953783708 hasRelatedWork W2791824431 @default.
- W2953783708 hasRelatedWork W2904174853 @default.
- W2953783708 hasRelatedWork W2951786554 @default.
- W2953783708 hasRelatedWork W2953783708 @default.
- W2953783708 hasRelatedWork W2963041618 @default.
- W2953783708 hasRelatedWork W3126776133 @default.
- W2953783708 hasRelatedWork W3186840088 @default.
- W2953783708 hasRelatedWork W4287064118 @default.
- W2953783708 hasRelatedWork W4287234591 @default.
- W2953783708 isParatext "false" @default.
- W2953783708 isRetracted "false" @default.