Detection and Mitigation of Rare Subclasses in Neural Network Classifiers

11/28/2019
by   Colin Paterson, et al.
13

Regions of high-dimensional input spaces that are underrepresented in training datasets reduce machine-learnt classifier performance, and may lead to corner cases and unwanted bias for classifiers used in decision making systems. When these regions belong to otherwise well-represented classes, their presence and negative impact are very hard to identify. We propose an approach for the detection and mitigation of such rare subclasses in neural network classifiers. The new approach is underpinned by an easy-to-compute commonality metric that supports the detection of rare subclasses, and comprises methods for reducing their impact during both model training and model exploitation.

READ FULL TEXT

page 6

page 7

page 8

page 11

page 12

research
07/07/2022

An Exploration of How Training Set Composition Bias in Machine Learning Affects Identifying Rare Objects

When training a machine learning classifier on data where one of the cla...
research
05/31/2023

Bias Mitigation Methods for Binary Classification Decision-Making Systems: Survey and Recommendations

Bias mitigation methods for binary classification decision-making system...
research
03/09/2023

Identification of Systematic Errors of Image Classifiers on Rare Subgroups

Despite excellent average-case performance of many image classifiers, th...
research
12/03/2018

Using Multitask Learning to Improve 12-Lead Electrocardiogram Classification

We develop a multi-task convolutional neural network (CNN) to classify m...
research
12/13/2021

Long-tail Recognition via Compositional Knowledge Transfer

In this work, we introduce a novel strategy for long-tail recognition th...
research
09/11/2022

A point process model for rare event detection

Detecting rare events, those defined to give rise to high impact but hav...
research
03/20/2013

Ensembling classification models based on phalanxes of variables with applications in drug discovery

Statistical detection of a rare class of objects in a two-class classifi...

Please sign up or login with your details

Forgot password? Click here to reset