FRIDAYS' AI TALKS - 12 May | Online

Fridays' AI talks are digital talks on emergent AI topics, organised by MSc in Artificial Intelligence. The series consist of talks given by scientists and researchers from academia and industry AI experts, and intends to cover the wide spectrum of topics regarding artificial intelligence. We plan to organize 1 talk per month and sporadically even including panel discussions. The target audience for the Fridays' AI talks includes the current MSc in Artificial Intelligence students, our alumni, members from academia, researchers and practitioners from labs and industry.

Fridays' AI talk: Neural Abstractive Summarization: Methods and Applications

Invited speaker: Grigorios Tsoumakas, Associate Professor of Machine Learning and Knowledge Discovery, School of Informatics, Aristotle University of Thessaloniki

Friday 15 May 2023, 18.00, Online

"This talk focuses on recent work of our team on abstractive summarization. After a brief introduction to modern neural summarization, I will present our divide-and-conquer approach for dealing with long documents and its application to summarizing scientific articles. I will then discuss Bayesian active summarization, our approach to combining active learning with state-of-the-art summarization models. Next, I will share our efforts towards controlling the output of summarization models given a particular context, such as a topic. Finally, I will briefly present applications in healthcare and finance".

Short bio

Grigorios Tsoumakas received a degree in Informatics from the Aristotle University of Thessaloniki (AUTH) in 1999, an MSc in Artificial Intelligence from the University of Edinburgh in 2000 and a PhD in Informatics from AUTH in 2005. He is an Associate Professor of Machine Learning and Knowledge Discovery at the School of Informatics of AUTH since 2020, where he has also served as Assistant Professor (2013 – 2020) and Lecturer (2007 – 2013).

His research expertise focuses on supervised machine learning (ensemble methods, multi-target prediction) and document intelligence (semantic indexing, keyphrase extraction, summarization). He has published more than 150 articles and according to Google Scholar his work has received more than 16k citations and his h-index is 50. His honors include receiving the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) 10-Year Test of Time Award in 2017 and the Marco Ramoni best paper award at the 19th International Conference on Artificial Intelligence in Medicine (AIME) 2021. He serves in the editorial board of Machine Learning and of Data Mining and Knowledge Discovery, as well as in the steering committee of ECML PKDD. Dr. Tsoumakas is a senior member of IEEE and ACM. He is an advocate of applied research that matters and has worked as a machine learning engineer, researcher and consultant in several national, international, and private sector funded R&D projects.

He is academic co-founder of Medoid AI, a spin-off company of AUTH established in 2019.

The recording will be soon available.