SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks

Nguyen, Van Nhan; Løkse, Sigurd; Wickstrøm, Kristoffer; Kampffmeyer, Michael; Roverso, Davide; Jenssen, Robert

doi:10.1007/978-3-030-58592-1_8

Cited by 26 publications

(16 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the above improvements on the feature embeddings, how to design a good similarity metric is also quite important for metric-based methods. Therefore, Nguyen et al [37] proposed the squared root of the Euclidean distance and the Norm distance (SEN) to enhance the discriminative ability of Euclidean distance. Wertheimer et al [38] proposed feature map reconstruction networks (FMPN) to reconstruct query features with respect to support features as the metric for non-parametric classifier.…”

Section: Meta-learning-based Methodsmentioning

confidence: 99%

Feature hallucination in hypersphere space for few‐shot classification

Yang

Fan

Chen

2022

IET Image Processing

View full text Add to dashboard Cite

Few‐shot classification (FSC) targeting at classifying unseen classes with few labelled samples is still a challenging task. Recent works show that transfer‐learning based approaches are competitive with meta‐learning ones, which usually pre‐train a convolutional neural networks (CNN)‐based network using cross‐entropy (CE) loss and throw away the last layer to post‐process the novel classes. Hereby, they still suffer the issue of getting a more transferable extractor and lacking enough labelled novel samples. Thus, the authors propose the algorithm of feature hallucination in hypersphere space (FHHS) for FSC. On the first stage, the authors pre‐train a more transferable feature extractor using a hypersphere loss (HL), which supplies CE with supervised contrastive (SC) loss and self‐supervised loss (SSL), in which SC can map the base and novel images onto the hypersphere space densely. On the second stage, the authors generate new samples for unseen classes using their novel algorithm of synthetic novel sampling with the base (SNSB), which linearly interpolate between each novel class prototype and its K nearest neighbour base class prototypes. Comprehensive experiments on multiple popular FSC demonstrate that HL loss can enhance the performance of backbone network and the authors’ feature hallucination method is superior to the existing hallucination‐based methods.

show abstract

Section: Meta-learning-based Methodsmentioning

confidence: 99%

Feature hallucination in hypersphere space for few‐shot classification

Yang

Fan

Chen

2022

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Variational scaling [62,2020] CNAPS [63,2020] TEAM [65,2020] SEN [66,2020] DeepEMD [68,2020] FSL with embedded class models [54,2019] Two-stage FSL [55,2020] TapNet [56,2019] FSL with global class representations [57,2019] KGTN [58,2020] Learning task-agnostic features Siamese Network [13,2015] Matching Network [9,2016] TPN [29,2019] PTN [30,2021] DN4 [32,2019] ATL-Net [35,2020] COMET [36,2021] PARN [33,2019] Attentive Prototype [37,2020] Cross-domain FSL [38,2020]…”

Section: Few-shot Deep Metric Learning Methodsmentioning

confidence: 99%

Section: Few-shot Deep Metric Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Revisiting metric learning for few-shot image classification

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

The goal of few-shot learning is to recognize new visual concepts with just a few amount of labeled samples in each class. Recent effective metric-based few-shot approaches employ neural networks to learn a feature similarity comparison between query and support examples. However, the importance of feature embedding, i.e., exploring the relationship among training samples, is neglected. In this work, we present a simple yet powerful baseline for few-shot classification by emphasizing the importance of feature embedding. Specifically, we revisit the classical triplet network from deep metric learning, and extend it into a deep K-tuplet network for few-shot learning, utilizing the relationship among the input samples to learn a general representation learning via episode-training. Once trained, our network is able to extract discriminative features for unseen novel categories and can be seamlessly incorporated with a non-linear distance metric function to facilitate the few-shot classification. Our result on the miniImageNet benchmark outperforms other metric-based few-shot classification methods. More importantly, when evaluated on completely different datasets (Caltech-101, CUB-200, Stanford Dogs and Cars) using the model trained with miniImageNet, our method significantly outperforms prior methods, demonstrating its superior capability to generalize to unseen classes.

show abstract

“…1) Metric-based approaches. This line of works focuses on learning a task-agnostic metric space and then predicting novel classes by a nearest-centroid classifier with Euclidean or cosine distance such as (Nguyen et al 2020;Snell, Swersky, and Zemel 2017;Li et al 2019b). 2) Optimizationbased approaches.…”

Section: Introductionmentioning

confidence: 99%

MetaNODE: Prototype Optimization as a Neural ODE for Few-Shot Learning

Zhang

et al. 2022

AAAI

View full text Add to dashboard Cite

Few-Shot Learning (FSL) is a challenging task, i.e., how to recognize novel classes with few examples? Pre-training based methods effectively tackle the problem by pre-training a feature extractor and then predicting novel classes via a cosine nearest neighbor classifier with mean-based prototypes. Nevertheless, due to the data scarcity, the mean-based prototypes are usually biased. In this paper, we attempt to diminish the prototype bias by regarding it as a prototype optimization problem. To this end, we propose a novel meta-learning based prototype optimization framework to rectify prototypes, i.e., introducing a meta-optimizer to optimize prototypes. Although the existing meta-optimizers can also be adapted to our framework, they all overlook a crucial gradient bias issue, i.e., the mean-based gradient estimation is also biased on sparse data. To address the issue, we regard the gradient and its flow as meta-knowledge and then propose a novel Neural Ordinary Differential Equation (ODE)-based meta-optimizer to polish prototypes, called MetaNODE. In this meta-optimizer, we first view the mean-based prototypes as initial prototypes, and then model the process of prototype optimization as continuous-time dynamics specified by a Neural ODE. A gradient flow inference network is carefully designed to learn to estimate the continuous gradient flow for prototype dynamics. Finally, the optimal prototypes can be obtained by solving the Neural ODE. Extensive experiments on miniImagenet, tieredImagenet, and CUB-200-2011 show the effectiveness of our method.

show abstract

SEN: A Novel Feature Normalization Dissimilarity Measure for Prototypical Few-Shot Learning Networks

Cited by 26 publications

References 9 publications

Feature hallucination in hypersphere space for few‐shot classification

Feature hallucination in hypersphere space for few‐shot classification

Revisiting metric learning for few-shot image classification

MetaNODE: Prototype Optimization as a Neural ODE for Few-Shot Learning

Contact Info

Product

Resources

About