Concept

Practical Convenience Causes Label-Correction Bias in Dev Sets

In a dev set of 1,000 examples with a classifier that has 98.0% accuracy, it is easier to examine the 20 misclassified examples than all 980 correctly classified examples. Because it is easier in practice to check only the misclassified examples, bias does creep into some dev sets.

0

1

Updated 2026-05-25

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Related