DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data

05/05/2021
by   Damien Dablain, et al.
99

Despite over two decades of progress, imbalanced data is still considered a significant challenge for contemporary machine learning models. Modern advances in deep learning have magnified the importance of the imbalanced data problem. The two main approaches to address this issue are based on loss function modifications and instance resampling. Instance sampling is typically based on Generative Adversarial Networks (GANs), which may suffer from mode collapse. Therefore, there is a need for an oversampling method that is specifically tailored to deep learning models, can work on raw images while preserving their properties, and is capable of generating high quality, artificial images that can enhance minority classes and balance the training set. We propose DeepSMOTE - a novel oversampling algorithm for deep learning models. It is simple, yet effective in its design. It consists of three major components: (i) an encoder/decoder framework; (ii) SMOTE-based oversampling; and (iii) a dedicated loss function that is enhanced with a penalty term. An important advantage of DeepSMOTE over GAN-based oversampling is that DeepSMOTE does not require a discriminator, and it generates high-quality artificial images that are both information-rich and suitable for visual inspection. DeepSMOTE code is publicly available at: https://github.com/dd1github/DeepSMOTE

READ FULL TEXT

page 1

page 9

page 10

page 11

research
04/05/2020

Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks

The fact that image datasets are often imbalanced poses an intense chall...
research
12/15/2022

Interpretable ML for Imbalanced Data

Deep learning models are being increasingly applied to imbalanced data i...
research
02/10/2023

DOMINO: Domain-aware Loss for Deep Learning Calibration

Deep learning has achieved the state-of-the-art performance across medic...
research
11/09/2020

MAGNeto: An Efficient Deep Learning Method for the Extractive Tags Summarization Problem

In this work, we study a new image annotation task named Extractive Tags...
research
04/14/2023

Toward Real-Time Image Annotation Using Marginalized Coupled Dictionary Learning

In most image retrieval systems, images include various high-level seman...
research
07/17/2019

Design and Evaluation of Product Aesthetics: A Human-Machine Hybrid Approach

Aesthetics are critically important to market acceptance in many product...
research
08/26/2021

Using GAN-based models to sentimental analysis on imbalanced datasets in education domain

While the whole world is still struggling with the COVID-19 pandemic, on...

Please sign up or login with your details

Forgot password? Click here to reset