What do the properties of neural tissue have to do with the way we process data in convolutional neural networks (CNNs)? Perhaps more than we think. This talk argues that, if anything, neural tissue teaches us that the hardware and software acceleration architectures we are currently building to support AI might not be the right ones, and maybe the solution is right under our nose.

About the Speaker

Nir Shavit received B.Sc. and M.Sc. degrees in Computer Science from the Technion – Israel Institute of Technology in 1984 and 1986, and a Ph.D. in Computer Science from the Hebrew University of Jerusalem in 1990. Shavit is a co-author of the book The Art of Multiprocessor Programming and a recipient of the 2004 Gödel Prize in theoretical computer science for his work on applying tools from algebraic topology to model shared memory computability and of the 2012 Dijkstra Prize in Distributed Computing for the introduction of Software Transactional Memory. He is a Professor of Computer Science at MIT and CEO of Neural Magic, a startup delivering GPU class performance of neural network execution on commodity CPUs.