Progress in autonomous vehicles and digital assistants shows that today’s machines can perform certain human tasks with remarkable accuracy. But they lack versatility and only work within limited domains for which large sets of training data are available. Transfer of knowledge, common sense, and an understanding of causality are open problems. As machine learning systems become pervasive, their role in our society is scrutinised and fascinating research areas open up, including accountability and fairness of automatic systems.

Bernhard Schölkopf is director at the Max Planck Institute for Intelligent Systems in Tübingen. With his foundational research, Bernhard has become one of the most important voices in international machine learning. His work includes methods of artificial intelligence that distinguish coincidental correlation from actual causation, as well as machine learning applications in fields ranging from medicine to astronomy. His research has won him several prestigious awards and he recently helped to start an AI lab funded by Amazon. At Falling Walls, Bernhard will show the essential role his research has in shaping the intelligent systems that are increasingly becoming part of our daily lives.