Image 01

BELIEF 2014
3rd International Conference on Belief Functions

 

               

       
Invited Speakers

We are delighted to announce the invited speakers at the BELIEF 2014 conference : Professor Thomas Lukasiewicz and Professor Nando de Freitas, both from the Computer Science department of Oxford University.



Prof. Thomas Lukasiewicz, Oxford University, UK

   

Thomas Lukasiewicz received the Dipl.-Inf. degree (M.Sc.) in Computer Science in 1993 from the Clausthal University of Technology, Germany, the Doctorate (Ph.D.) in Computer Science in 1996 from the University of Augsburg, Germany, and the Dozent degree (venia docendi) in Practical and Theoretical Computer Science in 2001 from the Vienna University of Technology, Austria. From 1993 to 1996, he was Research Assistant at the Institute of Computer Science of the University of Augsburg. From 1997 to 1999, he was Assistant Professor at the Institute of Computer Science of the University of Giessen. From 1999 to 2001, he was holding a Habilitation Fellowship by the German Research Foundation (DFG) at the Institute of Information Systems of the Vienna University of Technology. From 2001 to 2004, he was holding a Marie Curie Individual Fellowship by the European Union at the Department of Computer and System Sciences of the University of Rome "La Sapienza", Italy. From 2004 to 2009, he was holding a prestigious Heisenberg Fellowship (which is equivalent to an Associate Professorship) by the German Research Foundation (DFG): from 2004 to 2007 affiliated both at the Department of Computer and System Sciences of the University of Rome "La Sapienza" and at the Institute of Information Systems of the Vienna University of Technology, and from 2007 to 2009 affiliated both at the Department of Computer Science of the University of Oxford, UK, and at the Institute of Information Systems of the Vienna University of Technology. Since 2010, he is Professor of Computer Science and Yahoo! Research Fellow at the Department of Computer Science of the University of Oxford. He received the IJCAI-01 Distinguished Paper Award (best paper of 796 submitted and 197 accepted papers at the 17th International Joint Conference on Artificial Intelligence, Seattle, Washington, USA, August 4-10, 2001) for the paper Complexity Results for Structure-Based Causality (joint with Thomas Eiter) and the AIJ Prominent Paper Award 2013 (best paper published not more than five years ago in the Artificial Intelligence Journal (AIJ)) for the paper Combining Answer Set Programming with Description Logics for the Semantic Web (joint with Thomas Eiter, Giovambattista Ianni, Roman Schindlauer, and Hans Tompits).

Title of the talk: "Uncertainty in the Semantic Web"

Prof. Nando de Freitas, Oxford University, UK

   

I am a Professor of Computer Science at the University of Oxford. I was previously a full professor in machine learning and artificial intelligence at the department of computer science of the University of British Columbia. There, I was also an adjunct professor of statistics, and cognitive systems. I received my PhD from Trinity College, Cambridge University in 2000 on Bayesian methods for neural networks. From 1999 to 2001, I was a postdoctoral fellow at UC Berkeley in the artificial intelligence group of Stuart Russell.
I am a fellow of the Canadian Institute For Advanced Research (CIFAR) in the successful Neural Computation and Adaptive Perception program. I am an occasional entrepreneur: Zite, a big-data spin-off from my previous lab, has won numerous awards and was sold to CNN in 2011. Among my recent awards are aDistinguished Paper Award at the International Joint Conference on Artificial intelligence (IJCAI 2013), the 2012Charles A. McDowell Award for Excellence in Research, and the 2010 Mathematics of Information Technology and Complex Systems (MITACS) Young Researcher Award.

Title of the talk: "Deep Beliefs"

This talk entertains the view of encoding beliefs as continuous, deep, latent embeddings. I will discuss how contrastive learning approaches for probabilistic models relate to multi-task learning of deep embeddings for perceptual inputs and language. Subsequently, I will address the problems of inference, knowledge transfer, reasoning, imitation learning and decision making within this framework.