A
Alejandro Peña
Researcher at Autonomous University of Madrid
Publications - 10
Citations - 125
Alejandro Peña is an academic researcher from Autonomous University of Madrid. The author has contributed to research in topics: Facial expression & Unstructured data. The author has an hindex of 5, co-authored 10 publications receiving 66 citations.
Papers
More filters
Posted Content
InsideBias: Measuring Bias in Deep Networks and Application to Face Gender Biometrics.
TL;DR: This work analyzes how bias affects deep learning processes through a toy example using the MNIST database and a case study in gender detection from face images, and proposes InsideBias, a novel method to detect biased models.
Proceedings ArticleDOI
Bias in Multimodal AI: Testbed for Fair Automatic Recruitment
TL;DR: In this article, a fictitious automated recruitment testbed, FairCVtest, is proposed to evaluate the ability of the Artificial Intelligence (AI) behind such recruitment tool to extract sensitive information from unstructured data, and exploit it in combination to data biases in undesirable (unfair) ways.
Posted Content
Bias in Multimodal AI: Testbed for Fair Automatic Recruitment.
TL;DR: The methodology and results show how to generate fairer AI-based tools in general, and in particular fairer automated recruitment systems, as well as a list of recent works developing techniques capable of removing sensitive information from the decision-making process of deep learning architectures.
Proceedings ArticleDOI
InsideBias: Measuring Bias in Deep Networks and Application to Face Gender Biometrics
TL;DR: InsideBias as mentioned in this paper is a method to detect biased models based on how the models represent the information instead of how they perform, which is the normal practice in other existing methods for bias detection.
Proceedings ArticleDOI
Facial Expressions as a Vulnerability in Face Recognition
TL;DR: A huge facial expression bias in the most widely used databases, as well as a related impact of face expression in the performance of state-of-the-art algorithms are demonstrated.