Computer science at the service of the human brain or vice versa? The quest for understanding, simulating and exploring the world of brain computations.

by Dr Pawel Herman (Computational Biology and Neurocomputing, Computer Science and Communication, KTH)

Friday 12 Jun 2015, 13:00 → 14:00 Europe/Stockholm

Fantum (Fantum)

The human brain is commonly considered as the most complex structure known to mankind - some 100 billion neural cells organized into neural populations, intricate circuits, networks and systems via more than 100 trillion connecting synapses, and operating at a wide range of spatio-temporal scales, offer exceptional computational capabilities reflected in the human perception, memory, behaviour and, broadly speaking, impressive cognitive function. Despite heroic efforts over the last decades we can openly admit that our understanding of the underlying neural mechanisms is still very limited. Unravelling this mystery requires a multi-disciplinary endeavour and computer science has begun to play a very important role in this quest for knowledge. The first part of the seminar will briefly outline the wealth of contributions computer science has made and is going to make to brain science, within and beyond a rapidly expanding field of neuroinformatics. The focus will be on a computational modelling and simulation approach as a modern tool for integrating available biological data, exploring experimental as well as theoretical hypotheses and studying neural systems at various levels of physical and functional organisation. In addition, the emerging body of knowledge about neural computations that underlie the impressive array of brain capabilities, unparalleled in the known universe, offers a tempting opportunity to harness and mimic them to design intelligent systems. A vision that many aspects of the rich and robust computational functionality of the human brain can be reproduced in silico (on a computer) has captivated the imagination of scientists and stimulated progress in emerging fields of brain-like or neuro-inspired computing. The second part of the seminar will be devoted to the discussion of this type of synergies between brain and computer science. The acquired knowledge and intuition about principles of neural information processing can be effectively exploited to devise new unconventional computing frameworks and algorithms with far-reaching impact on computer science, robotics and information technology. In other words, the brain could serve as a fertile source of inspiration for novel computational paradigms capable of addressing particularly challenging real-world problems.