Mitigating confounding bias in recommendation via information bottleneck
How to effectively mitigate the bias of feedback in recommender systems is an important
research topic. In this paper, we first describe the generation process of the biased and …
research topic. In this paper, we first describe the generation process of the biased and …
Debiased representation learning in recommendation via information bottleneck
How to effectively mitigate the bias of feedback in recommender systems is an important
research topic. In this article, we first describe the generation process of the biased and …
research topic. In this article, we first describe the generation process of the biased and …
Is fairness only metric deep? evaluating and addressing subgroup gaps in deep metric learning
Deep metric learning (DML) enables learning with less supervision through its emphasis on
the similarity structure of representations. There has been much work on improving …
the similarity structure of representations. There has been much work on improving …
Fair representations by compression
X Gitiaux, H Rangwala - Proceedings of the AAAI Conference on …, 2021 - ojs.aaai.org
Organizations that collect and sell data face increasing scrutiny for the discriminatory use of
data. We propose a novel unsupervised approach to map data into a compressed binary …
data. We propose a novel unsupervised approach to map data into a compressed binary …
Identifying and disentangling spurious features in pretrained image representations
Neural networks employ spurious correlations in their predictions, resulting in decreased
performance when these correlations do not hold. Recent works suggest fixing pretrained …
performance when these correlations do not hold. Recent works suggest fixing pretrained …
Information-theoretic regularization for learning global features by sequential VAE
Sequential variational autoencoders (VAEs) with a global latent variable z have been
studied for disentangling the global features of data, which is useful for several downstream …
studied for disentangling the global features of data, which is useful for several downstream …
Robust classification under class-dependent domain shift
Investigation of machine learning algorithms robust to changes between the training and test
distributions is an active area of research. In this paper we explore a special type of dataset …
distributions is an active area of research. In this paper we explore a special type of dataset …
Pre-and Post-Fairness Processing for Black-Box Classifiers
X Gitiaux - 2022 - search.proquest.com
Abstract Machine learning algorithms increasingly support decision-making systems in
contexts where outcomes have long-term implications on the subject's well-being. At issue is …
contexts where outcomes have long-term implications on the subject's well-being. At issue is …
[图书][B] Improving Deep Representations by Incorporating Domain Knowledge and Modularization for Synthetic Aperture Radar and Physiological Data
T Agarwal - 2022 - search.proquest.com
Abstract Machine Learning (ML) using Artificial Neural Networks (ANNs), referred to as
Deep Learning (DL), is a very popular and powerful method of statistical inference. A …
Deep Learning (DL), is a very popular and powerful method of statistical inference. A …
[PDF][PDF] On the Tradeoff Between Accuracy and Fairness in Representation Learning
T Galstyan, H Khachatrian - РОССИЙСКО-АРМЯНСКОГО …, 2020 - vestnik.rau.am
In many applications of machine learning, it is desirable to have models which not only have
good accuracy on the prediction task but are also “fair” with respect to some protected …
good accuracy on the prediction task but are also “fair” with respect to some protected …