Pith. sign in

REVIEW 1 cited by

Beyond Size and Class Balance: Alpha as a New Dataset Quality Metric for Deep Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2407.15724 v2 pith:4BFVVQAH submitted 2024-07-22 cs.CV cs.LG

Beyond Size and Class Balance: Alpha as a New Dataset Quality Metric for Deep Learning

classification cs.CV cs.LG
keywords sizebalanceclassdatasetperformanceunicodex2013best
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

In deep learning, achieving high performance on image classification tasks requires diverse training sets. However, the current best practice$\unicode{x2013}$maximizing dataset size and class balance$\unicode{x2013}$does not guarantee dataset diversity. We hypothesized that, for a given model architecture, model performance can be improved by maximizing diversity more directly. To test this hypothesis, we introduce a comprehensive framework of diversity measures from ecology that generalizes familiar quantities like Shannon entropy by accounting for similarities among images. (Size and class balance emerge as special cases.) Analyzing thousands of subsets from seven medical datasets showed that the best correlates of performance were not size or class balance but $A$$\unicode{x2013}$"big alpha"$\unicode{x2013}$a set of generalized entropy measures interpreted as the effective number of image-class pairs in the dataset, after accounting for image similarities. One of these, $A_0$, explained 67% of the variance in balanced accuracy, vs. 54% for class balance and just 39% for size. The best pair of measures was size-plus-$A_1$ (79%), which outperformed size-plus-class-balance (74%). Subsets with the largest $A_0$ performed up to 16% better than those with the largest size (median improvement, 8%). We propose maximizing $A$ as a way to improve deep learning performance in medical imaging.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Scalably computing metric magnitude

    math.NA 2026-07 conditional novelty 5.0

    Hierarchical low-rank solvers beat dense and sparsified approaches for metric magnitude solves in experiments up to n=30,000, with a projected path to n≈10^5 via a containerized STRUMPACK/MPI pipeline.