DeepAI AI Chat
Log In Sign Up

MetaBalance: High-Performance Neural Networks for Class-Imbalanced Data

by   Arpit Bansal, et al.

Class-imbalanced data, in which some classes contain far more samples than others, is ubiquitous in real-world applications. Standard techniques for handling class-imbalance usually work by training on a re-weighted loss or on re-balanced data. Unfortunately, training overparameterized neural networks on such objectives causes rapid memorization of minority class data. To avoid this trap, we harness meta-learning, which uses both an ”outer-loop” and an ”inner-loop” loss, each of which may be balanced using different strategies. We evaluate our method, MetaBalance, on image classification, credit-card fraud detection, loan default prediction, and facial recognition tasks with severely imbalanced data, and we find that MetaBalance outperforms a wide array of popular re-sampling strategies.


page 1

page 2

page 3

page 4


Balanced-MixUp for Highly Imbalanced Medical Image Classification

Highly imbalanced datasets are ubiquitous in medical image classificatio...

Influence-Balanced Loss for Imbalanced Visual Classification

In this paper, we propose a balancing training method to address problem...

Neural Collapse Inspired Attraction-Repulsion-Balanced Loss for Imbalanced Learning

Class imbalance distribution widely exists in real-world engineering. Ho...

VOS: a Method for Variational Oversampling of Imbalanced Data

Class imbalanced datasets are common in real-world applications that ran...

KDE sampling for imbalanced class distribution

Imbalanced response variable distribution is not an uncommon occurrence ...

Credit risk prediction in an imbalanced social lending environment

Credit risk prediction is an effective way of evaluating whether a poten...

Cascaded Neural Networks with Selective Classifiers and its evaluation using Lung X-ray CT Images

Lung nodule detection is a class imbalanced problem because nodules are ...