Systems AI for Robot Learning

The Research Department Systems AI for Robot Learning (SAIROL) focuses on fundamental research on machine learning for intelligent autonomous robotic systems. This research agenda encompasses the development of methods in Systems AI for robot learning (in particular, the development of learning methods that enable model building, behavior generation, and information extraction in high-dimensional spaces, reinforcement learning, and can be used for exploration), architectures specifically for the application of machine learning for autonomous intelligent robot systems (e.g. for acquiring probabilistic forward models, learning and adapting problem-solving strategies, semi-autonomous human-robot interaction, and better use of sensing), and the use of learning robots in cognitive science (in particular, biologically and neurally-inspired approaches to AI, machine learning in brain-robot interfaces, and robot-based rehabilitation and prosthetics).