Title: Two tales of Applied AI: Recommendation Systems and Radiological Report Representation and Generation

Speaker: Prof. Denis Parra @ Pontifical Catholic University of Chile

Time : 11:00 ~ 12:00, Feb 11th, 2025

Location: Hybrid

• In-person: 26312

• Online: https://us02web.zoom.us/j/82598006604?pwd=cqBXBPbe3PFBTLrJYZjmRP5aqlINHY.1 

  • Meeting ID: 825 9800 6604

  • Passcode: 749926

Language: English speech & English slides

Abstract: 

In this talk, I will talk about two research topics related to my research on applied Artificial Intelligence: 1) recommendation systems, and 2) radiological report generation. In the first topic, I will describe an article recently published in the ICML conference where we investigate the unexpected effectiveness of reinforcement learning for sequential recommender systems, providing support on why it is indeed an unexpected result, and how to make sense of it. In the second part of this talk, I will talk about our group research on representing and generating radiological reports from medical images of chest X-rays. I will introduce the problem, current SOTA methods and challenges for future research. I will connect this latest problem with the development of Multimodal models.

Bio:

Denis Parra is Associate Professor at the Department of Computer Science, in the School of Engineering at the Pontifical Catholic University of Chile, one of the top universities in Latin America based on various rankings. Professor Parra holds a professional title of Informatics Engineering, class of 2004 from UACh, Valdivia, Chile; and a Ph.D. in Information Science from the University of Pittsburgh, USA, where he was supervised by Professor Peter Brusilovsky. 

Professor Parra is also affiliated with large national research centers in Chile. Notably, he is Principal Researcher at the Millenium Institute for Intelligent Healthcare Engineering (iHealth), as well as associate researcher at the National Center for Artificial Intelligence (CENIA). He has received awards such as the best student paper award at the UMAP conference in 2011, and he has been a candidate for best paper at the ACM Hypertext conference in 2013, as well as the ACM IUI conference in 2018 and 2019. He serves as senior program committee member in conferences such ACM RecSys (recommender systems) and ACM IUI (Intelligent User Interfaces). Prof. Parra's main research interests are Recommender Systems, and Applications of Machine Learning in Medical Imaging, Creative AI and Information Visualization.