Year
Month
(Preprint) Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder
Hanwen Liang ¹, Qiong Zhang 张琼 ², Peng Dai ¹, Juwei Lu ¹
¹ Huawei Noah’s Ark Lab, Canada
加拿大 华为诺亚方舟实验室
² Department of Statistics, University of British Columbia, Vancouver, Canada
arXiv, 2021-08-11
Abstract

State of the art (SOTA) few-shot learning (FSL) methods suffer significant performance drop in the presence of domain differences between source and target datasets. The strong discrimination ability on the source dataset does not necessarily translate to high classification accuracy on the target dataset.

In this work, we address this cross-domain few-shot learning (CDFSL) problem by boosting the generalization capability of the model. Specifically, we teach the model to capture broader variations of the feature distributions with a novel noise-enhanced supervised autoencoder (NSAE).

NSAE trains the model by jointly reconstructing inputs and predicting the labels of inputs as well as their reconstructed pairs. Theoretical analysis based on intra-class correlation (ICC) shows that the feature embeddings learned from NSAE have stronger discrimination and generalization abilities in the target domain. We also take advantage of NSAE structure and propose a two-step fine-tuning procedure that achieves better adaption and improves classification performance in the target domain.

Extensive experiments and ablation studies are conducted to demonstrate the effectiveness of the proposed method. Experimental results show that our proposed method consistently outperforms SOTA methods under various conditions.
Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder_1
Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder_2
Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder_3
Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencoder_4
  • Scaling metasurface design: from reusable unit cells to full-metasurface inverse design
  • Eunbi Jang, Jiwan Hong, Ho-Jun Lee, Seokju Choi, Hyewon Suk, Chanik Kang, Sunae So, Haejun Chung
  • Opto-Electronic Science
  • 2026-10-20
  • Ultrafast all-optical biomimetic encoder for spiking neural networks
  • Qijun Bian, Fangchen Hu, Junlian Jia, Aolong Sun, Sizhe Xing, Boyu Dong, An Yan, Ziwei Li, Jianyang Shi, Nan Chi, Junwen Zhang
  • Opto-Electronic Science
  • 2026-10-20
  • Windmill-shaped multi-unit piezoelectric ceramic tuning fork for cooperatively enhanced photoacoustic methane sensing
  • Botao Liu, Guowei Han, Runqiu Wang, Xuan Yang, Haiyue Sun, Shunda Qiao, Ying He, Angelo Sampaolo, Vincenzo Spagnolo, Yufei Ma
  • Opto-Electronic Science
  • 2026-10-20
  • Ultrafast laser direct writing of highly conductive copper micropatterns
  • Songyan Xue, Chunsan Deng, Mingduo Zhang, Huace Hu, Fayu Chen, Leimin Deng, Jianguo Liu, Hui Gao, Wei Xiong
  • Opto-Electronic Manufacturing
  • 2026-09-30
  • Recent advances in fiber-tip integration based on femtosecond laser 3D printing
  • Yuye Xuan, Dezhi Zhu, Dejun Liu, Ying Wang, Yiping Wang, Changrui Liao
  • Opto-Electronic Manufacturing
  • 2026-09-30
  • Neural holography enabled multi-beam steering for multi-task classifier
  • Lin Wu, Mian Wu, Zhen Li, Ying Qiu, Zichen Liu, Yongquan Zeng, Guozhen Liang, Hanbing Li, Guoxing Zheng, Zhixue He, Xi Xiao, Qi Jie Wang, Jin Tao, Shaohua Yu
  • Opto-Electronic Technology
  • 2026-09-30
  • Laser direct writing of nonlinear metasurfaces on phase-change material Ge₂Sb₂Te₅ for enhanced third harmonic generation
  • Lanpeng Guo, Haochong Liu, Shixin Gao, Ming Tang, Minghui Hong, Tun Cao
  • Opto-Electronic Technology
  • 2026-09-30
  • Nearly vertically emitted nonlinear light sources with high efficiencies enabled by high-Q nearly-degenerate polygon modes
  • Chuntao Li, Yingnuo Qiu, Renhong Gao, Lingling Qiao, Jielei Ni, Min Wang, Jintian Lin, Ya Cheng
  • Opto-Electronic Technology
  • 2026-09-30
  • Multiple degrees-of-freedom optical encryption via pixel-resolved vectorial holography
  • Jianzhong Peng, Haoxian Zhou, Mengzhe Shen, Dapeng Wang, Yu Chen, Renlong Zhang, Fangrui Lin, Xiaoyu Weng, Junle Qu
  • Opto-Electronic Technology
  • 2026-09-30
  • Recent progress on mid-infrared and terahertz dual-comb spectroscopy
  • Zhuoren Wan, Min Li, Binbin Liu, Yu Xia, Yuan Chen, Xiuxiu Zhang, Dongxu Zhu, Yan Dai, Sinan Tao, Hua Li, Ming Yan, Heping Zeng
  • Opto-Electronic Technology
  • 2026-09-30
  • Optical packaging of photonic integrated circuits: bottlenecks and potential solutions
  • Rodrigo Rendeiro, Hugo Neto, Diogo Vaz, Carla Rodrigues, Heming Wei, Jan Nedoma, Antonio Teixeira, Francisco Rodrigues, Carlos Marques
  • Opto-Electronic Technology
  • 2026-09-30



  • Retrieval & Interaction Machine for Tabular Data Prediction                                All-climate aqueous Na-ion batteries using “Water-in-Salt” electrolyte
    About
    |
    Contact
    |
    Copyright © PubCard