Few-shot learning via tensor hallucination

2021/04/19 by Michalis Lazarou, Lazarou, Michalis, Yannis Avrithis +3
Computer Science · Medicine · #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.2104.09467

openalex publication_date 2021/04/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Few-shot classification addresses the challenge of classifying examples given only limited labeled data. A powerful approach is to go beyond data augmentation, towards data synthesis. However, most of data augmentation/synthesis methods for few-shot classification are overly complex and sophisticated, e.g. training a wGAN with multiple regularizers or training a network to transfer latent diversities from known to novel classes. We make two contributions, namely we show that: (1) using a simple loss function is more than enough for training a feature generator in the few-shot setting; and (2) learning to generate tensor features instead of vector features is superior. Extensive experiments on miniImagenet, CUB and CIFAR-FS datasets show that our method sets a new state of the art, outperforming more sophisticated few-shot data augmentation methods.

Citations

• Few-Shot Learning with Graph Neural Networks
• Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning
• Learning Multiple Layers of Features from Tiny Images

Related

• Tensor feature hallucination for few-shot learning
• Label Hallucination for Few-Shot Classification
• A Closer Look at Few-shot Classification Again
• Adversarial Feature Hallucination Networks for Few-Shot Learning
• Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks