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

06/17/2021
by   Arpit Bansal, et al.
0

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.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

09/20/2021

Balanced-MixUp for Highly Imbalanced Medical Image Classification

Highly imbalanced datasets are ubiquitous in medical image classificatio...
10/06/2021

Influence-Balanced Loss for Imbalanced Visual Classification

In this paper, we propose a balancing training method to address problem...
09/29/2021

Multi-loss ensemble deep learning for chest X-ray classification

Class imbalance is common in medical image classification tasks, where t...
10/05/2020

Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance

Class-imbalance is one of the major challenges in real world datasets, w...
09/07/2018

VOS: a Method for Variational Oversampling of Imbalanced Data

Class imbalanced datasets are common in real-world applications that ran...
11/02/2017

Oversampling for Imbalanced Learning Based on K-Means and SMOTE

Learning from class-imbalanced data continues to be a common and challen...
04/28/2018

Credit risk prediction in an imbalanced social lending environment

Credit risk prediction is an effective way of evaluating whether a poten...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.