Imitating the brain, in silicon
Source:MIT
MIT researchers have created a semiconductor chip that is said to imitate how the brain’s neurons adapt in response to new information. With about 400 transistors, the silicon chip can simulate the activity of a single brain synapse — a connection between two neurons that allows information to flow from one to the other. The chip could also be used in neural prosthetic device research related to artificial retinas, says Chi-Sang Poon, a principal research scientist in the Harvard-MIT Division of Health Sciences and Technology.
Poon is the senior author of a paper describing the chip in the Proceedings of the National Academy of Sciences the week of Nov. 14. Guy Rachmuth, a former postdoc in Poon’s lab, is lead author of the paper. Other authors are Mark Bear, the Picower Professor of Neuroscience at MIT, and Harel Shouval of the University of Texas Medical School.
The researchers work relies on the their understanding of how a presynaptic neuron releases neurotransmitters, such as glutamate and GABA, which bind to receptors on a postsynaptic cell membrane, activating ion channels. Opening and closing those channels is said to change the cell’s electrical potential, sometimes strongly enough for the cell to fire an electrical impulse called an action potential.
“All of this synaptic activity depends on the ion channels, which control the flow of charged atoms such as sodium, potassium and calcium. Those channels are also key to two processes known as long-term potentiation (LTP) and long-term depression (LTD), which strengthen and weaken synapses, respectively,” according to an MIT press release.
The MIT researchers designed their computer chip so that the transistors could mimic the activity of different ion channels. This is an analog chip in which a gradient of electrical potential drives current to flow through the transistors just as ions are thought to flow through ion channels in a cell.
“We can tweak the parameters of the circuit to match specific ion channels,” Poon says. “We now have a way to capture each and every ionic process that’s going on in a neuron.”
The MIT researchers plan to use their chip to build systems to model specific neural functions, such as the visual processing system. Such systems could be much faster than digital computers.
Question: Do they actually understand how the brain works? Does this chip truly mimic that understanding? What good would it do if it did? What attempts failed before, and why?
Random thoughts on Google maps
As I headed out on the road for SearchSOA.com, I needed driving directions. Needed to find the bucolic burgh of, well let’s call it Medfordshire, Massachusetts. Where it is does not really matter.
To get driving directions, the evening previous to this excursion, I employed the great compute cloud known as Google. I asked this learned hand for instructions, and printed out a map that told me to proceed from my home in Boston’s Mission Hill, and take Route 95 North, etc.
Everything is going well. It is the next morning. Time to go. Alas, I’d left my Google Map print out back at the office. No problem, of course, I go online and get directions again.
After heading to Route 95 I pull out the directions and discover Google has computed a completely different set of plans to take me from and to the same end points. I put on my human hat and successfully made the trip. No problem. But it did get me to thinking about the compute cloud.
Clearly, somewhat random results work in some cases. But, a lot of enterprise computing must provide far more solid results. Read the rest of the story.
Cray Day! Make way for the PetaScale Jaguar
Where's Seymor?
It is an upgrade. But way up. How about a quadrillion mathematical calculations per second? Sounds prettier than ‘1.64 petaflops, n’cest pa? Yes, the folks at Oak Ridge have a hot-rod CPU. The makers call it Jaguar and mark it the world’s first petaflop system dedicated to open research. So apparently there’s more flops in the dark.
The upgrade at DOE’s Oak Ridge National Leadership Computing Facility represents a major milestone in a four-year project, begun in 2004 when DOE’s Office of Science launched a sustained effort to upgrade supercomputing capabilities for unclassified research at DOE’s complex of national laboratories.
Jaguar uses over 45,000 quad-core Opteron processors and features 362 terabytes of memory and a 10-petabyte file system. The machine has 578 terabytes per second of memory bandwidth and input/output (I/O) bandwidth of 284 gigabytes per second.
Cray is on the move on other fronts as well. They have just announced availability of NVIDIA Tesla C1060 GPU Computing processors in the new Cray CX1 line. Fits into a regular office without humongous refrigeration. Each Tesla processor has hundreds of processor cores that deliver nearly one teraflop of peak computing performance, we are told. It’s the revenge of the GPUs, one hopes.
Cyborg Google Krugman Luminary Nobel RS Future
Looking at Rolling Stone from last November. The special topic was The Future. For me it is back to The Future, just catching up with the reading. A number of luminaries and pseudo luminaries in the issue – Tim Berners-Lee, Bruce Springsteen, others. Everyone looks even less prescient a year later.
One I found of interest was Paul Krugman, who is by now even a more luminous luminary, having last month won a Nobel prize .
Economist and columnist Krugan seems to foreshadow recent events – he says what has been under way during the last four years is the dismantling of government. The collapse of credit markets for my money is proof of that.
Looking back: What is the most interesting thing he never saw coming. The pooling of human and machine intelligence.
Google is amazing. People don’t usually think of it in this way, but Google uses information technology to pool human judgment. It’s not the computer making decisions for us, but it’s not the computer working in a mechanical way, either – it’s computer enhanced intelligence. In an odd way, I guess you can say were all becoming cyborgs.
Wasn’t there not a type of human judgment pool in the preWeb world? That being the News. Did the editors control the pool more than they do now? In the heyday of the news, let’s say the late 19th Century, these editors were maniacally competing to sell papers, the decision of the public to pluck down two cents was the vote on the editor’s understanding of the pool’s notions. The pool spanned from the powerless [newly literate] up to the powerful. If you think today’s moguls and liberals control the news, you’d probably take the same view of Pulitzer…
Let’s not forget that the brilliance of the Google boys’ invention was based on a ready non-IT analog: The weighted reference merit system of learned articles. There was a judegment pool – though not a pool of a great large swath of humanity.
Getting back to Krugman’s future – and the question: Did the future turn out the way you imagined it when you were growing up? Krugman’s memories are much like mine, with the 1964 World’s Fair in prominence. I talked with a girl on the ATT World’s Fair Video phone from the Chicago Museum of Science and Industry and will never forget. And 2001, still top of the shelf at my pad. He mentions Asimov’s Foundation series as an inspiration. That I aint read. He says the books influenced him to be an economist although their goal was to tell the tale of psycho historians.
Note: This blog entry here seems as though it could appear on each of my web sites. It muses, so it could be on Moon Traveller Herald. It is on media so it could be Epitomime/Epitome. But it is here, on RJ-11 Amazing Techno Stories. Go configure.
Osamu Shimomura’s Jellyfish Nobel
Osamu Shimomura of Woods Hole Marine Biological Laboratory and B.U. was among three winners of the Nobel Prize for Chemistry. The others were Martin Chalfie and Roger Tsien. Each built on the work of the other to use the luminescent quality of fluorescent jelly fish proteins to create a method to observe the activity of biological cells.
Of old, people knew sea beings might glow in the dark. Shimomura and a colleague at Princeton in the 1960s identified the responsible green protein in Aequiorea Victoria Jellies, according to the New York Times. Chalfie of Coumbia U. saw use of the quality as a maker protein when spliced with the genes of a transparent round worm, Caenorhabditis elegans. Tsien mutated the green jelly fish proteins, and was able to create blue proteins, enabling observation of traces of multiple processes simultaneously. There is some weirdness here. Scientists have followed up on wthis work to create green glowing pigs.
Winning the Nobel disrupted 80-year-old Shimomura’s quietude. Caught by the Globe of Boston, he more or less said this is horrible. Shimomura came from Nagasaki and sas a youth survived the atomic bombing. He’s retired now, and has become a collector of art prints. “It’s important to have a hobby to make your mind correct and straight,” he said. “To avoid any misjudgment.” Me thinks he was born to blog!
Otellini, Intel and R&D: Good money after good
Seems for now Intel is on the upswing again. As appearing in this WSJ.com piece, Intel CEO Paul Otellini seems to have the Intel DNA. For the old guard, just the Intel building creates awe. When he took over, when things were bad Otellini put more into R&D. And good things happened. Nice material for the upcoming biopic on Otellini. At the heart of his process is the question ‘why?’
He learned this process from Mr. Grove: “Ask why, and ask it again five more times, until all of the artifice is stripped away and you end up with the intellectually honest answer.”
Still, I have been able to sit in on a few forums now on the topic of mutlicore programming. I don’t recall so much searching and stumbling, except maybe just before the web happened. Oh, yeah, the Transputer sessions seemed similarly muddled.
Multicore seems to be rife with cluelessness, and that could spell trouble for Intel, Moore’s Law or now Moore’s Law. Meanwhile, Ashlee Vance has appeared at NYT, and has a good take on Nvidia, which is the stalwart of the FPU, and also conjures for me the Anton Massively Parallel ASIC.
Fishnet cloak of negative refraction based invisibility
Let Me Feel Those Fishnet Hose
NYT and others report Berkeley researchers have used nano wires and fishnet, as described in Nature, to provide a cloak of invisibility. Based on the idea of negative refraction, the materials appear able to bend long-wavelength microwaves. A layered fishnet structure alternating between a metal and magnesium fluoride results in a metamaterial with a negative index of refraction for infrared light. Other Berkeley researchers in an almost concurrent publication in Science paper used a different approach, building an array of minuscule upright wires, which changed the electric fields of passing light waves. These both recall Retro-reflective Projection Technology (RPT) of 2003 from Tachi Lab. Which is maybe unweildy, but is still pretty much composable from off the shelf parts. In conversation with a poet colleague these advances were mentioned as portals perhaps to future powerful fields of poetic chorus. One recalls Gleem’s Invisible Shield too, and worries that Claude Rains’s ghost lurks. Worry Worry Worry.
Anton massive ASIC machine tackles the molecular
The unique requirements of protein simulation drive a new approach to supercomputers known as Anton. Named after microscope inventor Anton van Leeuwenhoek, the machine combines 512 ASIC parallel processors dedicated to the task of calculating 3D molecule interactions.
The machine was devised and sponsored, writes the NYTimes’ John Markoff, by D.E. Shaw, founder of D.E. Shaw & Co., an investment firm that specializes in computer-based trade modeling. The formal news of the new system, still in development, appeared in Communications of the ACM. [The latest issue of the ACM marks a very deliberate effort to revive the stalwart publication.]
Alternatives to Anton include IBM’s Gene/L supercomputer and the Folding@home [like Seti@home] project. Anton brings to light the hegemony of general-purpose supercomputing [read: massive off the shelf computing] that marked the end of the Cold War. Specialized and down-right weird computers were more prevalent before the latest run-up of Moore’s Law in the commercial sector. Back in the day there was a fairly wide variety of array processors, DSP accelerators, vector machines and so on. All outside the ken of CISCs and RISCs.
More extensive simulation may mean faster time to market for very important cancer trials. Or maybe not. But one hopes it means something good.
A Noble Award for IBM magneto-wizard Parkin
In case you missed it, IBM fellow Stuart Parkin was awarded the Daniel E. Noble Award for fundamental contributions to the development of magneto-resistive devices for nonvolatile RAM. These advances greatly enhanced the capabilities of modern disk drives. In recent years Parkin has been prominent in work that led to Racetrack Memory, which uses electron spin, rather than the charge, to create electronic devices. Parkin received the award along with Jim Daughton and Saied Tehrani. The award was announced at the IEEE International Magnetics Conference in Madrid.
Greg Leake’s .NET StockTrader 2.0 tries out load balancing
.NET StockTrader 2.0 shows how Microsoft thinks many people will use WCF to build SOA applications. It also has a bit of the look and feel of some Java app frameworks for distributed caching! If that is so it is a bit of a trojan horse.
