SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning

Yan Wang, Wei-Lun Chao, Kilian Q. Weinberger, Laurens van der Maaten
11/12/2019
cs.CV

Official code: mileyan/simple_shot (134 stars, MIT, last commit 2021-03-04).

Abstract

Few-shot learners aim to recognize new object classes based on a small number of labeled training examples. To prevent overfitting, state-of-the-art few-shot learners use meta-learning on convolutional-network features and perform classification using a nearest-neighbor classifier. This paper studies the accuracy of nearest-neighbor baselines without meta-learning. Surprisingly, we find simple feature transformations suffice to obtain competitive few-shot learning accuracies. For example, we find that a nearest-neighbor classifier used in combination with mean-subtraction and L2-normalization outperforms prior results in three out of five settings on the miniImageNet dataset.

View on arXivView PDF

Code Implementations(1)

mileyan/simple_shotOfficial100%
MIT
13427PythonNov 5, 20195 years agoMIT

Similar Papers

A Closer Look at Few-shot Classification

88% match1 repo1.2k2019

Meta-Learning with Differentiable Convex Optimization

88% match1 repo5432019

Prototypical Networks for Few-shot Learning

88% match1 repo1.2k2017

Meta-Tuning Loss Functions and Data Augmentation for Few-Shot Object Detection

87% match2023

What Makes for Effective Few-shot Point Cloud Classification?

86% match1 repo172023

Incremental Few-Shot Object Detection via Simple Fine-Tuning Approach

86% match1 repo102023

Cite this paper

@article{wang2019simpleshot,
  title  = {SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learning},
  author = {Yan Wang and Wei-Lun Chao and Kilian Q. Weinberger and Laurens van der Maaten},
  year   = {2019},
  eprint = {1911.04623},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1911.04623}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1911.04623.svg)](https://codeofpaper.com/paper/1911.04623)

Discussion