Despite decades of progress, today’s processors and AI accelerators still fall far short of biological brains in many respects. Neuromorphic chips, such as Intel’s Loihi, aim to narrow this gap by applying principles of brain computation, re-thinking computer architecture and algorithms from transistors up. This talk covers learnings from our research with Loihi and the outlook to commercialization of this technology.
Hosted by Microsystems Technology Laboratory.
Speaker
Mike Davies
Mike Davies is Director of Intel’s Neuromorphic Computing Lab. In the 2000s at Fulcrum Microsystems, Mike pioneered asynchronous design in industry-leading networking switches. Since 2015 his group in Intel Labs has been applying those methods to brain-inspired AI processors such as Loihi.
Explore
Discovering the Joy of Future-forward Electrical Engineering
Jane Halpern | Department of Electrical Engineering and Computer Science
One year in, MIT’s hands-on 6-5 (Electrical Engineering With Computing) degree program is already one of the most popular majors among first-year students.
New Method Could Increase LLM Training Efficiency
Adam Zewe | MIT News
By leveraging idle computing time, researchers can double the speed of model training while preserving accuracy.
Tomás Palacios appointed Director of ISN
Office of the Vice President for Research
As director, Tomás will lead ISN’s research mission and build communities within MIT and with external partners.