EnD: Entangling and Disentangling deep representations for bias correction

03/02/2021
by   Enzo Tartaglione, et al.
0

Artificial neural networks perform state-of-the-art in an ever-growing number of tasks, and nowadays they are used to solve an incredibly large variety of tasks. There are problems, like the presence of biases in the training data, which question the generalization capability of these models. In this work we propose EnD, a regularization strategy whose aim is to prevent deep models from learning unwanted biases. In particular, we insert an "information bottleneck" at a certain point of the deep neural network, where we disentangle the information about the bias, still letting the useful information for the training task forward-propagating in the rest of the model. One big advantage of EnD is that we do not require additional training complexity (like decoders or extra layers in the model), since it is a regularizer directly applied on the trained model. Our experiments show that EnD effectively improves the generalization on unbiased test sets, and it can be effectively applied on real-case scenarios, like removing hidden biases in the COVID-19 detection from radiographic images.

READ FULL TEXT

page 6

page 7

page 8

research
08/04/2020

A non-discriminatory approach to ethical deep learning

Artificial neural networks perform state-of-the-art in an ever-growing n...
research
12/01/2022

Task Discovery: Finding the Tasks that Neural Networks Generalize on

When developing deep learning models, we usually decide what task we wan...
research
09/30/2022

Information Removal at the bottleneck in Deep Neural Networks

Deep learning models are nowadays broadly deployed to solve an incredibl...
research
04/26/2022

Unsupervised Learning of Unbiased Visual Representations

Deep neural networks are known for their inability to learn robust repre...
research
05/05/2023

Mining bias-target Alignment from Voronoi Cells

Despite significant research efforts, deep neural networks are still vul...
research
06/10/2023

Revealing Model Biases: Assessing Deep Neural Networks via Recovered Sample Analysis

This paper proposes a straightforward and cost-effective approach to ass...
research
02/21/2022

Photometric Redshift Estimation with Convolutional Neural Networks and Galaxy Images: A Case Study of Resolving Biases in Data-Driven Methods

Deep Learning models have been increasingly exploited in astrophysical s...

Please sign up or login with your details

Forgot password? Click here to reset