WhiteBox – LOEWE Research Cluster on explainable models for AI launched

The Centre for Cognitive Science successfully applied for the LOEWE Research Cluster “WhiteBox – explainable models for human and artificial intelligence”. The cluster will be funded four years with a total grant of 4.7 Mio EUR by the Hessian State Ministry of Higher Education, Research and the Arts. WhiteBox is joining the twin disciplines AI and Cognitive Science. January 2021, an interdisciplinary team of 9 PIs and their teams will start investigating the project's core questions: How can we better explain artificial and human intelligence?

Artificial intelligence (AI) is a central technology of the 21st century. Self-driving cars, medical diagnostic systems, intelligent robots, and smart assistants promise an increase in automation of tasks, that until recently could only be carried out by humans. Recently, AI systems have surpassed human intelligence in some specific tasks, e.g. at the games of chess or poker. Many of the recent successes in AI rely on artificial deep neural networks. These systems are inspired by the human brain and are trained to find statistical regularities in large data ensembles.

But, their predictions and decisions have turned out to be inexplicable even to the researchers who built these systems. This is unfortunate, since for many economic or societal applications the availability of explainable and reliable predictions and decisions are indispensable.

WhiteBox is aimed at developing methods at the intersection between Cognitive Science and AI to make human and artificial intelligence more understandable.

More detailed information can be found on the project's webpage.