Catching Image Retrieval Generalization

Maksim Zhdanov, Ivan Karpukhin
6/23/2023
cs.LGcs.CV

No confident code implementation found for this paper.

Abstract

The concepts of overfitting and generalization are vital for evaluating machine learning models. In this work, we show that the popular Recall@K metric depends on the number of classes in the dataset, which limits its ability to estimate generalization. To fix this issue, we propose a new metric, which measures retrieval performance, and, unlike Recall@K, estimates generalization. We apply the proposed metric to popular image retrieval methods and provide new insights about deep metric learning generalization.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Loss Functions and Metrics in Deep Learning

81% match2023

Variance & Greediness: A comparative study of metric-learning losses

81% match2026

Model-agnostic Measure of Generalization Difficulty

80% match1 repo32023

A practical generalization metric for deep networks benchmarking

80% match2024

Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets

80% match1 repo4.3k2022

Machine Learning vs Deep Learning: The Generalization Problem

80% match2024

Cite this paper

@article{zhdanov2023catching,
  title  = {Catching Image Retrieval Generalization},
  author = {Maksim Zhdanov and Ivan Karpukhin},
  year   = {2023},
  eprint = {2306.13357},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2306.13357v1}
}

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/2306.13357.svg)](https://codeofpaper.com/paper/2306.13357)

Discussion