School of Computing and Communications Distinguished Seminar: Machine Reading with No Supervision
Tuesday 1 October 2019, 12:30pm to 1:30pm
Venue
Faraday Lecture Theatre, Bailrigg, Lancaster, LA1 4YWOpen to
Alumni, Applicants, External Organisations, Postgraduates, Prospective Undergraduate Students, Public, Staff, UndergraduatesRegistration
Registration not required - just turn upTicket Price
The Seminar Series is open to all and free to attend, so please feel free to come along.Event Details
The School of Computing and Communications will have a distinguished seminar given by Dr. Sebastian Riedel, who is Professor of Natural Language Processing and Machine Learning at UCL and a researcher at Facebook AI research.
In this seminar Prof. Sebastian Riedel (UCL & Facebook AI Research) will give a brief overview of Facebook AI Research (FAIR), and then focus on recent work done in teaching machines to read with no supervision. In particular, he will show how to develop models that can answer questions in natural language, about natural language text, without the need of any example questions and answers.
About the speaker:
Prof. Sebastian Riedel currently is a Researcher at Facebook AI Research. He is also the Professor of Natural Language Processing and Machine Learning at University College London and an Allen Distinguished Investigator, leading the Machine Reading Lab. Before that, Prof. Riedel was a postdoc and research scientist at UMass Amherst, a researcher at Tokyo University and DBCLS. He got his Ph.D. from the University of Edinburgh.
Prof. Sebastian Riedel is generally interested in deep learning, good old graphical models as well as old-school symbolic AI. Specifically, He works on teaching machines how to read and reason, in the intersection of Natural Language Processing (NLP) and Machine Learning. Recently, Prof. Riedel has been tackling these problems via forms of end-to-end differentiable program interpreters as well as adversarial regularisation. You can find more details on Professor Riedel's work on Google Scholar.
Everyone is welcome!
Contact Details
Name | Wenjie Ruan |