ReMix: Calibrated Resampling for Class Imbalance in Deep learning

12/03/2020
by   Colin Bellinger, et al.
0

Class imbalance is a problem of significant importance in applied deep learning where trained models are exploited for decision support and automated decisions in critical areas such as health and medicine, transportation, and finance. The challenge of learning deep models from imbalanced training data remains high, and the state-of-the-art solutions are typically data dependent and primarily focused on image data. Real-world imbalanced classification problems, however, are much more diverse thus necessitating a general solution that can be applied to tabular, image and text data. In this paper, we propose ReMix, a training technique that leverages batch resampling, instance mixing and soft-labels to enable the induction of robust deep models for imbalanced learning. Our results show that dense nets and CNNs trained with ReMix generally outperform the alternatives according to the g-mean and are better calibrated according to the balanced Brier score.

READ FULL TEXT
research
06/29/2020

Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization

Real-world large-scale datasets are heteroskedastic and imbalanced – lab...
research
10/30/2018

Weak-supervision for Deep Representation Learning under Class Imbalance

Class imbalance is a pervasive issue among classification models includi...
research
04/19/2022

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

Class imbalance distribution widely exists in real-world engineering. Ho...
research
01/06/2020

Identifying and Compensating for Feature Deviation in Imbalanced Deep Learning

We investigate learning a ConvNet classifier with class-imbalanced data....
research
04/25/2017

Deep Over-sampling Framework for Classifying Imbalanced Data

Class imbalance is a challenging issue in practical classification probl...
research
05/24/2023

Mixture of Experts with Uncertainty Voting for Imbalanced Deep Regression Problems

Data imbalance is ubiquitous when applying machine learning to real-worl...

Please sign up or login with your details

Forgot password? Click here to reset