Computer Vision News - February 2017

Tsvi Achler is the founder and CEO of Optimizing Mind . He invented a ground-breaking brain- motivated AI core algorithm (a new approach both for learning and recognition), based on his research as a neuroscientist. Besides having engineering degrees from UC Berkeley , Tsvi is also an MD and has a PhD in Neuroscience from University of Illinois at Urbana-Champaign . Tsvi kindly accepted this fascinating talk with Computer Vision News . What did he find about our brain? Tsvi, can you tell us more about yourself and your background? From an early age, I think around 14, I was programming Assembly. After that I went to Berkeley to study electrical engineering and computer science. I became very interested in neuroscience and wanted to understand how the brain works, but from an algorithm perspective, from a techie perspective. Did you have any hope of understanding how the brain works? Well, hope or not, this is what I want to do [ interviewer laughs ]. What do you know from your studies about the brain that we don’t know? I don’t think anyone knows exactly what is going on in the brain, but the brain is not working like current “feedforward” artificial neural networks. I think the brain is extensively using feedforward-feedback connections. From an algorithm perspective, the key ideas of my theory are: one, that the feedforward- feedback connections are performing an optimization; two, that this optimization occurs during recognition and not during learning as feedforward networks suggest. Optimization generates error signals. I’m getting very technical here, but evidence of optimization during recognition and error signals can be observed as the most prevalent finding when one records from neurons: network-wide neuron bursting. [ continues at page 26 ] Computer Vision News Guest 23 Tsvi Achler, Optimizing Mind Guest “ There is so much in the brain that I want to know! ”