loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Authors: Michael Danner 1 ; Bakir Hadžić 2 ; Robert Radloff 2 ; Xueping Su 3 ; Leping Peng 4 ; Thomas Weber 2 and Matthias Rätsch 2

Affiliations: 1 CVSSP, University of Surrey, Guildford, U.K. ; 2 ViSiR, Reutlingen University, Germany ; 3 School of Electronics and Information, Xi’an Polytechnic University, China ; 4 Hunan University of Science and Technology, China

Keyword(s): Unbiased Machine Learning, Fairness, Trustworthy AI, Acceptance Research, Debiasing Training Data, Facial Data Sets, AI-Acceptance Analysis.

Abstract: AI-based prediction and recommender systems are widely used in various industry sectors. However, general acceptance of AI-enabled systems is still widely uninvestigated. Therefore, firstly we conducted a survey with 559 respondents. Findings suggested that AI-enabled systems should be fair, transparent, consider personality traits and perform tasks efficiently. Secondly, we developed a system for the Facial Beauty Prediction (FBP) benchmark that automatically evaluates facial attractiveness. As our previous experiments have proven, these results are usually highly correlated with human ratings. Consequently they also reflect human bias in annotations. An upcoming challenge for scientists is to provide training data and AI algorithms that can withstand distorted information. In this work, we introduce AntiDiscriminationNet (ADN), a superior attractiveness prediction network. We propose a new method to generate an unbiased convolutional neural network (CNN) to improve the fairn ess of machine learning in facial dataset. To train unbiased networks we generate synthetic images and weight training data for anti-discrimination assessments towards different ethnicities. Additionally, we introduce an approach with entropy penalty terms to reduce the bias of our CNN. Our research provides insights in how to train and build fair machine learning models for facial image analysis by minimising implicit biases. Our AntiDiscriminationNet finally outperforms all competitors in the FBP benchmark by achieving a Pearson correlation coefficient of PCC = 0.9601. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 18.191.186.84

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Danner, M. ; Hadžić, B. ; Radloff, R. ; Su, X. ; Peng, L. ; Weber, T. and Rätsch, M. (2023). Overcome Ethnic Discrimination with Unbiased Machine Learning for Facial Data Sets. In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP; ISBN 978-989-758-634-7; ISSN 2184-4321, SciTePress, pages 464-471. DOI: 10.5220/0011624900003417

@conference{visapp23,
author={Michael Danner and Bakir Hadžić and Robert Radloff and Xueping Su and Leping Peng and Thomas Weber and Matthias Rätsch},
title={Overcome Ethnic Discrimination with Unbiased Machine Learning for Facial Data Sets},
booktitle={Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP},
year={2023},
pages={464-471},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011624900003417},
isbn={978-989-758-634-7},
issn={2184-4321},
}

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP
TI - Overcome Ethnic Discrimination with Unbiased Machine Learning for Facial Data Sets
SN - 978-989-758-634-7
IS - 2184-4321
AU - Danner, M.
AU - Hadžić, B.
AU - Radloff, R.
AU - Su, X.
AU - Peng, L.
AU - Weber, T.
AU - Rätsch, M.
PY - 2023
SP - 464
EP - 471
DO - 10.5220/0011624900003417
PB - SciTePress