This brief essay, originated by the work on the Neuromorphics Lab in the DARPA SyNAPSE project, describes our early effort in the study of alternative computing schemes that will make use of massive memristive-based devices coupled with low-power CMOS processes to efficiently compute neural activation and learning in novel computing devices. The answer was to couple fuzzy inference with dense memristive memory. This combination can provide extensive power and silicon real estate savings while maintaining a high degree of accuracy in the resulting precision of the computations. Read the rest of this entry »
-
Fuzzy logic and memristive hardware
| August 9, 2011 -
Silicon brains
| July 19, 2011
This article, appeared on 7/19/2011 on AZoRobotics, discusses the main reasons why I believe that we are on the verge of a paradigm shift in the way robots are going to be programmed. AZoRobotics.com is part of AZoNetwork, a leading online science, engineering and medical publisher serving over 2.5 million monthly visitor sessions across its suite of sites.... so it's a good venue to get some valuable feedback.
Comments: Leave a commentAlso tagged neuromorphic technology, robot -
Memristors – a revolutionary research field
| December 19, 2010
In 1971, Prof. Leon Chua from UC Berkeley published an article about the memristor, a fourth passive element in electronics (in addition to resistor, capacitor and inductor). Since Chua's article, memristors have become an exciting device that potentially can revolutionize the electronic industry. Chua predicted, only because of symmetry reasons, that there is a missing passive element, whose main characteristic is a relationship between flux (the total voltage being applied on it) and electric charge (the total current that had flown through it). Such a device actually acts as a resistor, but unlike the conventional resistor, its resistance is not constant, nor depends upon the temporary voltage being applied to it (as is the case for non-linear resistors). Read the rest of this entry »
-
MoNETA IEEE article: a summary until now
| December 5, 2010
The online version of the IEEE Spectrum article describing our work on the MoNETA project (withing the DARPA SyNAPSE grant) has been out for a bit more than a week, and the story is starting to generate many comments and being picked up by blogs and magazines (see Slashdot and Popular Science). May be it's time to summarize what is happening, and starting to address the many comments related to the article. Ben and I will start a series of posts on the topic, this one being the first. Read the rest of this entry » -
Neurdons on IEEE Spectrum cover page!
| November 23, 2010
December won't be a boring month for Neurdons. Two of us (who is writing, and Ben Chandler) will be featured in the cover page, back page story, and in an article title "MoNETA: A Mind Made from Memristors. DARPA's new memristor-based approach to AI consists of a chip that mimics how neurons process information" on the December issue of IEEE Spectrum. The preview of the article has appeared online. Read the rest of this entry » -
Boston University Team SyNAPSE presentation, August DARPA site visit
| September 8, 2010
This post contains the videos of the Boston University team presentation during the recent DARPA SyNAPSE site visit occurred in August 2010, in Palo Alto. This presentation talks about the ongoing projects undertaken by the Boston University team in SyNAPSE since March 2010, with some of our most recent results and plans for the near future. Read the rest of this entry » -
HP and Hynix
| September 1, 2010
It was less than 24 hours ago since the last post echoing an article on the NYT on memristors, and two more articles have appeared on both the NYT and CNN with the reports of the announcement, made on Tuesday by HP, that it would commercialize a new computer memory technology with Hynix, the South Korean chip maker. Read the rest of this entry » -
Shrinking hopes
| August 31, 2010
It was 1965 when Gordon Moore (co-founder of Intel) made the observation that the number of transistors per square inch of integrated circuits roughly doubled each year, starting from the time of introduction of these devices. What was in 1965 an observation become a law postulating that this trend would not come to a stop. Unfortunately, the limits of physics have caused a recent slowdown in the rhythm at which semiconductor companies can miniaturize transistors. Yesterday, an article on the NYT focused on a fresh announcement by Rice University and HP on a fundamental discovery able to overcome the barrier to the continued miniaturization of electronic components. Read the rest of this entry » -
Leon Chua visits the Boston University SyNAPSE team
| June 2, 2010
Leo Chua gave one of his brilliant talks on May 21, 2010 at the ICCNS 2010 conference at the Department of Cognitive and Neural Systems. Just to be sure you get the magnitude of the event: Leon Chua is the first to theorize the existence of memristors with his paper "Memristor: The Missing Circuit Element" on IEEE TRANSACTIONS on Circuit Theory, published in September 1971. Read the rest of this entry » -
Beyond cats
| April 19, 2010
One of the main motivation of the Neurdon blog is to shed light on the true innovative force behind the introduction of new computing paradigms based on the nervous system functioning. This is why when neurdons read the word cat associated with large-scale simulations, we get a little sad... readers should know that the greatest revolution that will see AI and the semiconductor industry jointly turn upside down our current use of intelligent machines has nothing to do with a bunch of cats. Though, according to PHYSORG article "Cat brain: A step toward the electronic equivalent", the pinnacle of memristor-based computer evolution will be a cat. IBM's "Max" (no, not Max Versace... apparently my shortened name is widely used to name unaware felines across the globe...) will be in good company. Read the rest of this entry »
