Distribution Estimation Based Pseudo-Feature Library Generation For Few-Shot Image Classification

Zhang, Jiacheng; Liu, Weishuo; Zhao, Zhicheng; Su, Fei

doi:10.1109/icmew53276.2021.9455950

Cited by 4 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The other generation-based methods (Yang et al 2021;Zhang et al 2021) transfer the information from the similar base classes to the few-shot classes. Yang et al (Yang et al 2021) propose a distribution calibration strategy to calibrate the biased feature distributions of few-shot classes to the corresponding ground-truth distributions.…”

Section: Related Workmentioning

confidence: 99%

Feature Distribution Fitting with Direction-Driven Weighting for Few-Shot Images Classification

Wei

Wan

et al. 2023

AAAI

View full text Add to dashboard Cite

Few-shot learning has received increasing attention and witnessed significant advances in recent years. However, most of the few-shot learning methods focus on the optimization of training process, and the learning of metric and sample generating networks. They ignore the importance of learning the ground-truth feature distributions of few-shot classes. This paper proposes a direction-driven weighting method to make the feature distributions of few-shot classes precisely fit the ground-truth distributions. The learned feature distributions can generate an unlimited number of training samples for the few-shot classes to avoid overfitting. Specifically, the proposed method consists of two optimization strategies. The direction-driven strategy is for capturing more complete direction information that can describe the feature distributions. The similarity-weighting strategy is proposed to estimate the impact of different classes in the fitting procedure and assign corresponding weights. Our method outperforms the current state-of-the-art performance by an average of 3% for 1-shot on standard few-shot learning benchmarks like miniImageNet, CIFAR-FS, and CUB. The excellent performance and compelling visualization show that our method can more accurately estimate the ground-truth distributions.

show abstract

Section: Related Workmentioning

confidence: 99%

Feature Distribution Fitting with Direction-Driven Weighting for Few-Shot Images Classification

Wei

Wan

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…Representation learning, on the other hand, aims at identifying feature transformations which can allow simple statistical techniques like nearest neighbor and bayesian classification generalize on few-shot tasks for novel classes [7,44]. Starting from early works [25] and building to recent methodologies [45,48], representation learning strategies have relied on simple statistical assumptions that may not hold across diverse datasets.…”

Section: Introductionmentioning

confidence: 99%

GDC- Generalized Distribution Calibration for Few-Shot Learning

Kumar¹,

Zaidi²

2022

Preprint

View full text Add to dashboard Cite

Few shot learning is an important problem in machine learning as large labelled datasets take considerable time and effort to assemble. Most few-shot learning algorithms suffer from one of two limitationsthey either require the design of sophisticated models and loss functions, thus hampering interpretability; or employ statistical techniques but make assumptions that may not hold across different datasets or features. Developing on recent work in extrapolating distributions of small sample classes from the most similar larger classes, we propose a Generalized sampling method that learns to estimate few-shot distributions for classification as weighted random variables of all large classes. We use a form of covariance shrinkage to provide robustness against singular covariances due to overparameterized features or small datasets. We show that our sampled points are close to few-shot classes even in cases when there are no similar large classes in the training set. Our method works with arbitrary off-the-shelf feature extractors and outperforms existing state-of-the-art on miniImagenet, CUB and Stanford Dogs datasets by 3% to 5% on 5way-1shot and 5way-5shot tasks and by 1% in challenging cross domain tasks.

show abstract