Pablo Barros
Senior Research Scientist as Sony, External Professor at the University of Pernambuco

Brussels, Belgium
Recife, Brazil
Welcome to my personal website! I am a research scientist with over a decade of experience in developing technologies in social signal processing, intelligent agents, generative AI, and affective computing. My expertise centers around utilizing artificial neural networks to enhance human experiences through advancements in affective computing, social signal processing, and human-robot interaction.
Throughout my career, I have been committed to designing, training, and evaluating machine learning systems that improve the understanding and modeling of human behavior in automated systems. I have advised students at various academic levels (BSc, MSc, PhD) and continue to find joy in mentoring.
In addition to my main roles, I am an external professor at the University of Pernambuco, Brazil, where I am part of the Time Visão research group. I advise students, deliver lectures, and participate in innovative projects across many related fields.
This site showcases the projects that have marked my career—from my academic tenure, achieving a PhD and continuing as a postdoctoral researcher at the University of Hamburg, in Germany, to my contributions as a researcher at the Italian Institute of Technology, in Genova, Italy. Currently, I am part of the Digital Profile team at Sony, driving innovation and merging scientific rigor with strategic business applications.
Explore the site to discover public repositories I’ve contributed to and summaries of innovative projects that have defined my professional journey.
news
Jan 05, 2025 | Aceitando Alunos de Mestrado e Doutorado - Brazil |
---|---|
Dec 24, 2024 | Best Poster Awards at LACORO 2024 |
selected publications
- A personalized affective memory model for improving emotion recognitionIn International Conference on Machine Learning, 2019
- The omg-empathy dataset: Evaluating the impact of affective behavior in storytellingIn 2019 8th International Conference on Affective Computing and Intelligent Interaction (ACII), 2019
- Learning from learners: Adapting reinforcement learning agents to be competitive in a card gameIn 2020 25th International Conference on Pattern Recognition (ICPR), 2021
- Adapting the interplay between personalized and generalized affect recognition based on an unsupervised neural frameworkIEEE Transactions on Affective Computing, 2020
- Incorporating rivalry in reinforcement learning for a competitive gameNeural Computing and Applications, 2023
- Ciao! a contrastive adaptation mechanism for non-universal facial expression recognitionIn 2022 10th International Conference on Affective Computing and Intelligent Interaction (ACII), 2022