FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3047009786> ?p ?o ?g. }

• W3047009786 abstract "With the memory-resource-limited constraints, class-incremental learning (CIL) usually suffers from the catastrophic forgetting problem when updating the joint classification model on the arrival of newly added classes. To cope with the forgetting problem, many CIL methods transfer the knowledge of old classes by preserving some exemplar samples into the size-constrained memory buffer. To utilize the memory buffer more efficiently, we propose to keep more auxiliary low-fidelity exemplar samples rather than the original real high-fidelity exemplar samples. Such a memory-efficient exemplar preserving scheme makes the old-class knowledge transfer more effective. However, the low-fidelity exemplar samples are often distributed in a different domain away from that of the original exemplar samples, that is, a domain shift. To alleviate this problem, we propose a duplet learning scheme that seeks to construct domain-compatible feature extractors and classifiers, which greatly narrows down the above domain gap. As a result, these low-fidelity auxiliary exemplar samples have the ability to moderately replace the original exemplar samples with a lower memory cost. In addition, we present a robust classifier adaptation scheme, which further refines the biased classifier (learned with the samples containing distillation label knowledge about old classes) with the help of the samples of pure true class labels. Experimental results demonstrate the effectiveness of this work against the state-of-the-art approaches." @default.
• W3047009786 created "2020-08-10" @default.
• W3047009786 creator A5004882141 @default.
• W3047009786 creator A5007599551 @default.
• W3047009786 creator A5056078581 @default.
• W3047009786 creator A5064284315 @default.
• W3047009786 creator A5091841056 @default.
• W3047009786 date "2020-08-04" @default.
• W3047009786 modified "2023-09-23" @default.
• W3047009786 title "Memory Efficient Class-Incremental Learning for Image Classification" @default.
• W3047009786 cites W1594039573 @default.
• W3047009786 cites W1682403713 @default.
• W3047009786 cites W1979631293 @default.
• W3047009786 cites W1997865285 @default.
• W3047009786 cites W2048868165 @default.
• W3047009786 cites W2050453099 @default.
• W3047009786 cites W2060277733 @default.
• W3047009786 cites W2061879449 @default.
• W3047009786 cites W2110569316 @default.
• W3047009786 cites W2115403315 @default.
• W3047009786 cites W2130370504 @default.
• W3047009786 cites W2163605009 @default.
• W3047009786 cites W2187089797 @default.
• W3047009786 cites W2194775991 @default.
• W3047009786 cites W2317413155 @default.
• W3047009786 cites W2342650393 @default.
• W3047009786 cites W2473930607 @default.
• W3047009786 cites W2551868040 @default.
• W3047009786 cites W2554616628 @default.
• W3047009786 cites W2560647685 @default.
• W3047009786 cites W2591909298 @default.
• W3047009786 cites W2593768305 @default.
• W3047009786 cites W2605043629 @default.
• W3047009786 cites W2609698233 @default.
• W3047009786 cites W2611997167 @default.
• W3047009786 cites W2737492962 @default.
• W3047009786 cites W2738226240 @default.
• W3047009786 cites W2786446225 @default.
• W3047009786 cites W2786498526 @default.
• W3047009786 cites W2788388592 @default.
• W3047009786 cites W2791091755 @default.
• W3047009786 cites W2798869704 @default.
• W3047009786 cites W2801477643 @default.
• W3047009786 cites W2884282566 @default.
• W3047009786 cites W2885561447 @default.
• W3047009786 cites W2890126432 @default.
• W3047009786 cites W2890648518 @default.
• W3047009786 cites W2892936772 @default.
• W3047009786 cites W2892946488 @default.
• W3047009786 cites W2894094671 @default.
• W3047009786 cites W2898503115 @default.
• W3047009786 cites W2902625698 @default.
• W3047009786 cites W2918288062 @default.
• W3047009786 cites W2924448532 @default.
• W3047009786 cites W2925005147 @default.
• W3047009786 cites W2925804722 @default.
• W3047009786 cites W2948336998 @default.
• W3047009786 cites W2948734064 @default.
• W3047009786 cites W2950557191 @default.
• W3047009786 cites W2954929116 @default.
• W3047009786 cites W2955547856 @default.
• W3047009786 cites W2962884963 @default.
• W3047009786 cites W2962966271 @default.
• W3047009786 cites W2963003887 @default.
• W3047009786 cites W2963038864 @default.
• W3047009786 cites W2963311299 @default.
• W3047009786 cites W2963438784 @default.
• W3047009786 cites W2963559848 @default.
• W3047009786 cites W2963788399 @default.
• W3047009786 cites W2963813679 @default.
• W3047009786 cites W2963850662 @default.
• W3047009786 cites W2964067969 @default.
• W3047009786 cites W2964088867 @default.
• W3047009786 cites W2964151081 @default.
• W3047009786 cites W2964189064 @default.
• W3047009786 cites W2965407115 @default.
• W3047009786 cites W2966730026 @default.
• W3047009786 cites W2970567995 @default.
• W3047009786 cites W2972740417 @default.
• W3047009786 cites W2974317861 @default.
• W3047009786 cites W2981864462 @default.
• W3047009786 cites W2982220706 @default.
• W3047009786 cites W2995139074 @default.
• W3047009786 cites W2997654015 @default.
• W3047009786 cites W3007580291 @default.
• W3047009786 cites W3013325675 @default.
• W3047009786 cites W3018638193 @default.
• W3047009786 cites W3024202154 @default.
• W3047009786 cites W3035480894 @default.
• W3047009786 cites W3035519852 @default.
• W3047009786 cites W3103800629 @default.
• W3047009786 cites W3118608800 @default.
• W3047009786 cites W3141815693 @default.
• W3047009786 cites W3184035966 @default.
• W3047009786 doi "https://doi.org/10.48550/arxiv.2008.01411" @default.
• W3047009786 hasPublicationYear "2020" @default.
• W3047009786 type Work @default.
• W3047009786 sameAs 3047009786 @default.
• W3047009786 citedByCount "0" @default.
• W3047009786 crossrefType "posted-content" @default.

• next