FSL-QuickBoost: Minimal-Cost Ensemble for Few-Shot Learning
Name
3664647.3681446.pdf
Size
1.92 MB
Format
Adobe PDF
Checksum (MD5)
2f9c21028901afa8e6a4d59ebafeb25a
Author(s)
Bai, Yunwei
Cai, Bill Yang
Tan, Ying Kiat
Zheng, Zangwei
Chen, Shiming
Chen, Tsuhan
Date Issued
October 28, 2024
Publisher
ACM|Proceedings of the 32nd ACM International Conference on Multimedia
Citation
Bai, Yunwei, Cai, Bill Yang, Tan, Ying Kiat, Zheng, Zangwei, Chen, Shiming et al. 2024. "FSL-QuickBoost: Minimal-Cost Ensemble for Few-Shot Learning."
Version
Final published version
Abstract
Few-shot learning (FSL) usually trains models on data from one set of classes, but tests them on data from a different set of classes, providing a few labeled support samples of the unseen classes as a reference for the trained model. Due to the lack of target-relevant training data, there is usually high generalization error with respect to the test classes. In this work, we conduct empirical explorations and propose an ensemble method (namely QuickBoost), which is efficient and effective for improving the generalization of FSL. Specifically, QuickBoost includes an alternative-architecture pretrained encoder with a one-vs-all binary classifier (namely FSL-Forest) based on random forest algorithm, and is ensembled with the off-the-shelf FSL models via logit-level averaging. Experiments on three benchmarks demonstrate that our method achieves state-of-the-art performance with good efficiency. Codes are available at https://github.com/WendyBaiYunwei/FSL-QuickBoost.
Description
MM ’24, October 28-November 1, 2024, Melbourne, VIC, Australia
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3664647.3681446
