Few-Shot Lifelong Learning

被引:0
|
作者
Mazumder, Pratik [1 ]
Singh, Pravendra
Rai, Piyush [1 ]
机构
[1] IIT Kanpur, Dept Comp Sci & Engn, Kanpur, Uttar Pradesh, India
来源
THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE | 2021年 / 35卷
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Many real-world classification problems often have classes with very few labeled training samples. Moreover, all possible classes may not be initially available for training, and may be given incrementally. Deep learning models need to deal with this two-fold problem in order to perform well in real-life situations. In this paper, we propose a novel Few-Shot Lifelong Learning (FSLL) method that enables deep learning models to perform lifelong/continual learning on few-shot data. Our method selects very few parameters from the model for training every new set of classes instead of training the full model. This helps in preventing overfitting. We choose the few parameters from the model in such a way that only the currently unimportant parameters get selected. By keeping the important parameters in the model intact, our approach minimizes catastrophic forgetting. Furthermore, we minimize the cosine similarity between the new and the old class prototypes in order to maximize their separation, thereby improving the classification performance. We also show that integrating our method with self-supervision improves the model performance significantly. We experimentally show that our method significantly outperforms existing methods on the miniImageNet, CIFAR-100, and CUB-200 datasets. Specifically, we outperform the state-of-the-art method by an absolute margin of 19.27% for the CUB dataset.
引用
收藏
页码:2337 / 2345
页数:9
相关论文
共 50 条
  • [1] Few-Shot Few-Shot Learning and the role of Spatial Attention
    Lifchitz, Yann
    Avrithis, Yannis
    Picard, Sylvaine
    2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2021, : 2693 - 2700
  • [2] Lifelong ensemble learning based on multiple representations for few-shot object recognition
    Kasaei, Hamidreza
    Xiong, Songsong
    ROBOTICS AND AUTONOMOUS SYSTEMS, 2024, 174
  • [3] Learn Continually, Generalize Rapidly: Lifelong Knowledge Accumulation for Few-shot Learning
    Jin, Xisen
    Lin, Bill Yuchen
    Rostami, Mohammad
    Ren, Xiang
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, EMNLP 2021, 2021, : 714 - 729
  • [4] Federated Few-shot Learning
    Wang, Song
    Fu, Xingbo
    Ding, Kaize
    Chen, Chen
    Chen, Huiyuan
    Li, Jundong
    PROCEEDINGS OF THE 29TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2023, 2023, : 2374 - 2385
  • [5] Defensive Few-Shot Learning
    Li, Wenbin
    Wang, Lei
    Zhang, Xingxing
    Qi, Lei
    Huo, Jing
    Gao, Yang
    Luo, Jiebo
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (05) : 5649 - 5667
  • [6] Fractal Few-Shot Learning
    Zhou, Fobao
    Huang, Wenkai
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 35 (11) : 1 - 15
  • [7] Survey on Few-shot Learning
    Zhao K.-L.
    Jin X.-L.
    Wang Y.-Z.
    Ruan Jian Xue Bao/Journal of Software, 2021, 32 (02): : 349 - 369
  • [8] Variational Few-Shot Learning
    Zhang, Jian
    Zhao, Chenglong
    Ni, Bingbing
    Xu, Minghao
    Yang, Xiaokang
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 1685 - 1694
  • [9] Interventional Few-Shot Learning
    Yue, Zhongqi
    Zhang, Hanwang
    Sun, Qianru
    Hua, Xian-Sheng
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020, 2020, 33
  • [10] Memristive Cosine-Similarity-Based Few-Shot Learning with Lifelong Memory Adaptation
    Zhou, Houji
    Yang, Ling
    Bao, Han
    Li, Jiancong
    Li, Yi
    Miao, Xiangshui
    ADVANCED INTELLIGENT SYSTEMS, 2023, 5 (02)