Date: April 9
Hour: 1 p.m.
Venue: PA2 amphitheatre, Técnico – Alameda Campus
Speaker: Pedro Neto (FEUP e INESC TEC)
Title: “Ethical Issues in Face Recognition: Explainable AI and Synthetic Datasets”
Abstract:
The outstanding performance gains in Face Recognition over the years have been sustained by three main factors. 1) Computational power that enabled deep neural networks; 2) Deeper and more complex network architectures; 3) Large datasets containing different identities and images. Due to these three points (and a few tweaks on the loss function) it was possible to surpass humans in this task. Yet, the latter two points brought concerning problems with them. The complexity of the architectures translated into black-box systems that cannot be fully understood nor explained. And the collection of the datasets has raised concerns regarding the collection method, consent and ethical use of that data. Additionally, it seems that models trained on these datasets are frequently unfair towards gender and ethnic groups. To mitigate these problems, we cover three topics: 1) How to adapt explainable artificial intelligence to face recognition?; 2) Why are my models unfair and what can I do about it? Is it all about balancing the number of samples per ethnicity?; 3) Are we already capable of creating fully synthetic datasets for face recognition? Do models perform well when trained in these datasets?. There are state-of-the-art approaches tackling each of these problems, from a new taxonomy for xAI to diffusion models and data balancing considerations (applied to our solution at the CVPR 2024 FRCSyn competition).
Speaker bio:
Pedro C. Neto is a PhD candidate (at his final months) in Electrical and Computer Engineering at Faculdade de Engenharia da Universidade do Porto (FEUP) and a research assistant at Instituto de Engenharia, Sistemas e Computadores, Tecnologia e Ciência (INESC TEC) where he is associated with the Visual Computing and Machine Intelligence group. Currently, Pedro is also an Invited Teaching Assistant at FEUP/FCUP, where he teaches courses related to Machine Learning and Algorithm Design. His research focuses on the understanding and mitigation of biases in Face Recognition, and on explainability tools for the same task. He holds a MSc degree with honors in Computer Science awarded by Aalto University, Finland, and a BSc degree with honors in Software/Informatics Engineering awarded by the Instituto Superior de Engenharia do Porto (ISEP).
Priberam is a member of the IST Spin-Off Community®.
The Priberam Machine Learning Lunch Seminars are free of charge. Prior registration is required.
