CZS supported AI projects – session 2
13:20 AM – 14:15 PM
The algorithmic challenges of probabilistic inference are far from being solved and often become a bottleneck for probabilistic modeling and inference. Although probabilistic programming languages automate inference of models at various degrees of sophistication, the generic inferences of these languages are often outperformed by customized inference techniques. As of today, general-purpose inference remains very challenging. We will discuss how address these challenges by means of enabling technologies such as algorithm engineering, high-performance computing, automatic differentiation, and logic.
Speaker and Project leader
Joachim Giesen is a professor of Theoretical Computer Science at University of Jena since 2008. His research interests are in algorithms for machine learning and artificial intelligence. Before coming to Jena, his research focus was in the area of computational geometry. For two years, he worked at the Max Planck Institute for Computer
Science in Saarbrücken as a research group leader in the Max Planck Center for Visual Computing and Communication.
Before that, he was a PhD student, PostDoc, and senior researcher in the Theory of Combinatorial Algorithms group of Emo Welzl at ETH Zurich, and a PostDoc in Tamal K. Dey’s group at The Ohio State University. He obtained his habilitation in 2006 and his doctorate in 2000 from ETH Zurich.
Learning from experience and making predictions that will guide future actions are at the core of intelligence. These tasks need to embrace uncertainty to avoid the risk of drawing wrong conclusions or making bad decisions. The sources of uncertainty are manifold ranging from measurement noise, missing information, and insufficient data to uncertainty about good parameter values or the adequacy of a model. Probability theory offers a framework to represent uncertainty in the form of probabilistic models. The rules of probability theory allow us to manipulate and integrate uncertain evidence in a consistent manner. Based on these rules, we can make inferences about the world in the context of a given model. In this sense, probability theory can be viewed as an extended logic.