Plenary Speaker I
Prof. Dr. Tom Heskes
Radboud University Nijmegen, Netherlands
Speech Title: Causal Discovery from Big Data
Abstract: Discovering causal relations from data lies at the heart of most scientific research today. In apparent contradiction with the adagio “correlation does not imply causation”, recent theoretical insights indicate that such causal knowledge can also be derived from purely observational data, instead of only from controlled experimentation. In the “big data” era, such observational data is abundant and being able to actually derive causal relationships from very large data sets would open up a wealth of opportunities for improving business, science, government, and healthcare. In this talk, I will sketch how insights from statistics and machine learning may lead to novel approaches for robust discovery of relevant causal relationships with applications in health and genomics.
Tom Heskes is full professor of Artificial Intelligence. After receiving his PhD on neural networks, he worked as a postdoc at the Beckman Institute in Champaign-Urbana, Illinois. Back in the Netherlands, he joined SNN, the Foundation for Neural Networks. Tom Heskes is VICI laureate, received two TOP grants (EW and ZonMW), and has been (co-)leading various other national and European projects.
Heskes’ research concerns the development, understanding, and application of machine learning methods, currently in particular deep learning and causal inference. He works on applications in other scientific disciplines, in collaboration with physicists, astronomers, neuroscientists, biologists, and medical specialists, as well as in industry, among others through his spin-off company Machine2Learn.
Plenary Speaker II
Prof. Peter Andras
School of Computing and Mathematics, Keele University, UK
Speech Title: Trust, privacy and intelligent control
Abstract: Intelligent control is widespread today and is used in cars, home appliances and factory robots among other things. The advent of deep learning made possible to develop successful applications in many areas where classification, regression or pattern recognition is required on the basis of data. However, the success of these applications often masks unexplored uncertainties due to complex models of the data, naïve assumptions of the users, and unexpected situations of use. In order for devices relying on intelligent control to become fully and reliably accepted and to realize their potential benefits, it is important to understand how human trust relates to machines equipped with intelligent control. A particular aspect of trust relates to guaranteeing the privacy of the users, who let these machines access their behavioral data. This talk will address these issues, discuss illustrative examples and indicate possible directions of development for trustworthy and privacy preserving machines with intelligent control.
Professor Peter Andras has a BSc in computer science (1995), an MSc in artificial intelligence (1996) and a PhD in mathematical analysis of neural networks (2000), all from the Babes-Bolyai University, Cluj, Romania. He is a Professor in the School of Computing and Mathematics, Keele University, UK. He has published 2 books and over 100 papers. He works in the areas of machine learning, complex systems and computational neuroscience. Professor Andras is senior member of the IEEE, member of the International Neural Network Society (INNS), of the Society for Artificial Intelligence and Simulation of Behaviour (AISB), of the International Society for Artificial Life (ISAL) and fellow of the Royal Society of Biology (RSB).