Results 341 to 350 of about 842,651 (374)

Class-incremental learning with causal relational replay

Expert Systems with Applications
In Class-Incremental Learning (Class-IL), deep neural networks often fail to learn a sequence of classes incrementally due to catastrophic forgetting, a phenomenon arising from the absence of exposure to old knowledge. To alleviate this issue, conventional rehearsal methods, such as experience replay, store a limited number of old exemplars and then ...
Toan Nguyen 0004, Duc Kieu, Bao Duong, Tung Kieu, Kien Do, Thin Nguyen, Bac Le   +6 more
openaire   +2 more sources

Benchmarking Class Incremental Learning in Deep Learning Traffic Classification

IEEE Transactions on Network and Service Management
Traffic Classification (TC) is experiencing a renewed interest, fostered by the growing popularity of Deep Learning (DL) approaches. In exchange for their proved effectiveness, DL models are characterized by a computationally-intensive training procedure
Giampaolo Bovenzi, Alfredo Nascita, Lixuan Yang, A. Finamore, Giuseppe Aceto, D. Ciuonzo, A. Pescapé, Dario Rossi   +7 more
semanticscholar   +1 more source

Double distillation for class incremental learning

2021 23rd International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC), 2021
Darian M. Onchis, Ioan-Valentin Samuila
openaire   +1 more source

Improved Continually Evolved Classifiers for Few-Shot Class-Incremental Learning

IEEE transactions on circuits and systems for video technology (Print)
Few-shot class-incremental learning (FSCIL) aims to continually learn new classes using a few samples while not forgetting the old classes. The scarcity of new training data will seriously destroy the model’s stability and plasticity. Continually Evolved
Ye Wang, Yaxiong Wang, Guoshuai Zhao, Xueming Qian   +3 more
semanticscholar   +1 more source

Brain-inspired Class Incremental Learning

2022 IEEE 5th International Conference on Information Systems and Computer Aided Education (ICISCAE), 2022
Wei Wang, Zhiying Zhang, Jielong Guo
openaire   +1 more source

Few-Shot Class-Incremental Learning With Adjustable Pseudo-Incremental Sessions for Bearing Fault Diagnosis

IEEE Sensors Journal
Rotating machinery may constantly generate new classes of faults in complex operating environments, with a finite set of fault samples that are obtainable.
Hongyan Zhu, Changqing Shen, Jiaan Wang, Bojian Chen, Dong Wang, Zhongkui Zhu   +5 more
semanticscholar   +1 more source

Class-Incremental Learning: Survey and Performance Evaluation on Image Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023
Marc Masana, Xialei, Bartłomiej Twardowski   +2 more
exaly  

Exemplar-free class incremental learning via discriminative and comparable parallel one-class classifiers

Pattern Recognition, 2023
Wenju Sun, Yangli-Ao Geng
exaly  

Recurrent Network Expansion for Class Incremental Learning

IEEE Transactions on Neural Networks and Learning Systems
Class incremental learning (CIL) is the key to achieving adaptive vision intelligence, and one of the main streams for CIL is network expansion (NE). However, state-of-the-art (SOTA) methods usually suffer from feature diffusion, growing parameters, feature confusion, and classifier bias.
Kai Jiang, Xueru Bai, Feng Zhou
openaire   +2 more sources

Learning to Classify With Incremental New Class

IEEE Transactions on Neural Networks and Learning Systems, 2022
Da-Wei Zhou, Yang Yang, De-Chuan Zhan
exaly