How to Create a Mind, Brain Projects

Human Brain Project gets go-ahead, €1.19 billion over 10 years – 2013-01-28 12:00:00

The European Commission has announced the selection of the Human Brain Project (HBP) as one of two new flagship science projects. The project will last 10 years and is expected to cost €1.19 billion. Half of the funding will come from the European Commission itself, the other half will come from the EU member countries and participating institutions. Around 80 institutions are involved. Together they will attempt to model the human brain and build a biologically-realistic simulation of it within a supercomputer. The project will start around September 2013 and full brain simulations are targeted for 2023.

Ray Kurzweil joins Google as a research director – 2012-12-15 12:00:00

Ray Kurzweil announced yesterday that he’s joining Google as a research director. He will work full time at the Google HQ in California. His work will focus on machine learning and language processing. Last month Kuzweil published a book titled How to Create a Mind. In turn, Google’s long-term aim is to build a machine that is AI complete. So it makes perfect sense that Kuzweil and Google should team up.

Spaun – world’s largest functional brain model – 2012-11-29 12:00:00

Spaun is a simulated brain that contains 2.5 million neurons. This is far fewer than the 86 billion in the human brain, but enough to recognize lists of numbers, do simple arithmetic, and solve reasoning problems. The system is biologically realistic in its simulation of spiking neurons and neurotransmitters. As a result, it reproduces many quirks of human behaviour, such as the tendency to remember items at the start and end of a list better than those in the middle.

IBM simulation of 530 billion neurons and 137 trillion synapses – 2012-11-14 12:00:00

Using 96 Blue Gene/Q racks of the LLNL Sequoia supercomputer (1,572,864 processor cores, 1.5 PB memory), IBM and LBNL achieved a simulation of 2 billion neurosynaptic cores. This simulation contained an unprecedented 530 billion neurons and 137 trillion synapses running only 1542x slower than real time. The attached image represents the simulated network of pathways between neuron clusters in macaque monkey brain.

Blue Brain Project – Year Three documentary video published – 2012-11-13 17:35:00

The third part of Noah Hutton’s excellent documentary about the Blue Brain Project was released today. This is a ten-part film, with one part being released every year. It follows the project team as they attempt to build a complete simulation of the human brain within a supercomputer. All parts can be viewed on the website: bluebrainfilm.com

Ray Kurzweil’s new book published: How to Create a Mind – 2012-11-13 13:40:00

This is a probably a must-read for anyone interested in how artificial general intelligence will be achieved. His basic premise is that the brain contains no hidden secrets and that the creation of sentient machines will follow from re-engineering the brain in non-biological form. Kurzweil estimates that the human neocortex contains 300 million pattern processors linked horizontally and vertically. He claims it is these processors, rather than the neurons of which they are composed, that are the fundamental units of the brain. Near-future computer technology could build a synthetic brain containing well beyond “a mere 300 million” processors, and instead as many as a billion or a trillion. Website: howtocreateamind.com

Brain simulations and the new TOP500 supercomputer list – 2012-11-13 12:00:00

Twice a year a list of world’s the top 500 supercomputers is released. Today the list was updated and the new Titan supercomputer in the US took top slot at 17.59 petaflops. Rising to 5th place, with 4.141 petaflops, is the JuQUEEN machine in Jülich, Germany. If the Human Brain Project is successful in securing EU funding in January 2013 then the brain simulations will be run on JuQUEEN. This machine is now around 100x more powerful than the supercomputer currently used by the Blue Brain Project in Lausanne, Switzerland. JuQUEEN (pictured here) should be powerful enough to simulate all 100 million neurons of a mouse brain.

Google publish a neural network research paper – 2012-06-26 12:00:00

Google announce they have built a 9-layer neural network that can learn to detect faces using only unlabeled images. The model has 1 billion connections and the dataset has 10 million 200×200 pixel images randomly downloaded from the internet. It was trained on a cluster of 1,000 machines (16,000 cores) for three days.

Brain simulations and the new TOP500 supercomputer list – 2012-06-18 12:00:00

The list of the world’s top 500 supercomputers was updated today. At 8th place, with 1.38 petaflops, was the new JuQUEEN machine in Jülich, Germany. If the Human Brain Project is successful in securing EU funding in coming months then it’s likely that brain simulations will be run on JuQUEEN. This machine should be powerful enough to simulate all 100 million neurons of a mouse brain.

BrainScaleS neuromorphic processors – new demo video – 2012-05-31 12:00:00

This ten-minute video tour released last week shows the recently fully assembled BrainScaleS machine. We get a close look at the neuromorphic silicon wafer and all the supporting circuitry. There is then a demonstration of one artificial spiking neuron triggering the firing of a second neuron.

New video by IBM’s Brain Lab – 2012-05-30 12:00:00

This five-minute video released today gives some insight into IBM Research’s efforts to help shape the new age of cognitive computing. Project manager Dharmendra Modha gives us a quick tour of the Almaden lab in California and demonstrates some of the results to date. We see the so-called ‘brain wall’ which visualises 64 million neurons, equal to the number in a rat cerebral cortex. We also see the neuromorphic processor recognising hand-written digits and learning to play a game of pong.

SpikeFun biological neural network simulator for PCs – version 0.70 released – 2012-05-17 12:00:00

SpikeFun is a large-scale, biologically-realistic, neural network simulator that runs on a standard Windows PC. It can simulate around 32k neurons and 1.8 million synapses at 0.25x real-time on a normal home computer. On a high-end PC simulations of 3 million neurons and 476 million synapses have been achieved, although these were at 0.001x slower than real-time. Today version 0.70 was released and is available for free download.

Larry Page is personally funding worm brain emulation research – 2012-04-18 12:00:00

David Dalrymple, a neuroscientist at Harvard, recently gave a guest lecture at Marvin Minsky’s Society of Mind artificial intelligence class. David talked about his current research efforts to reverse engineer the C. elegans nervous system. During the Q&A session someone asked how the research is funded. To the surprise and amusement of the audience, David answered that it’s personally financed by Larry Page (CEO of Google). The research was also Larry’s idea.

Memristor-based artificial synapses built by HRL Labs – 2012-03-23 12:00:00

A press release from HRL Labs today says they have demonstrated the first functioning memristor array stacked on a conventional CMOS semiconductor circuit. They say this hybrid circuit is a critical advance in developing intelligent machines. Ultimately the team plans to scale the neuromorphic chip to support millions of neurons and billions of artificial synapses. The paper, which was actually published in December, is titled: “A functional hybrid memristor crossbar-array/CMOS system for data storage and neuromorphic applications”. The research is funded by the DARPA SyNAPSE program.

Web-based “Worm Brain/Body Browser” released – 2012-03-21 12:00:00

OpenWorm is an attempt to build a complete cellular-level simulation of the nematode worm Caenorhabditis elegans. Of the 959 cells in the hermaphrodite, 302 are neurons and 95 are muscles. The simulation will model neuromuscular electrical activity as well as physical forces within the worm and from its environment. The project is treated a first step towards simulating whole biological systems including, ultimately, the human brain.

Digital brain in the works at Qualcomm – 2012-02-15 12:00:00

Paul Jacobs, the CEO of Qualcomm, reveals an interesting update: “The team started out building a retina. They found it responded to optical illusions the same way a human does. They added another layer of cells and it started to find features. They added another layer, it started to find corners and oriented lines. Another layer, it started to find patterns. Today it tracks objects. It’s not programmed, it’s taught.”

Artificial Brains: Not in this century – 2012-02-04 12:00:00

this skeptical article was published yesterday by cognitive scientist Mark Changizi. One of his arguments is that we still don’t understand the tiny nervous system of the Caenorhabditis elegans roundworm. If the functioning of these 302 neurons is beyond our grasp, even after decades of study, then we’ve no hope of soon understanding the many billions of neurons in a mammalian brain.

BrainScaleS machine shows its first spiking neural activity – 2012-01-24 12:00:00

BrainScaleS is a European research project that is building neuromorphic hardware using wafer-scale integration. One wafer is designed to simulate ~50 million synapses, or up to 200,000 neurons. The wafer was delivered to Germany from the fab in Taiwan late 2011. Last week in the lab it displayed its first spiking neural activity. The wafer can be seen in the photo, encased behind an octagonal aluminium plate.

Human Brain Project now half-way through the one-year pilot phase – 2012-01-03 12:00:00

The proceedings of the 2nd European Future Technologies Conference and Exhibition were published online today. This conference marked the half-way point in the one-year pilot phase of the Human Brain Project (HBP). In the second half of 2012 the European Union will decide, via the FET Flagships program, whether to award .1 billion in funding for the project. If awarded, the HBP will become a ten-year attempt (2013 to 2023) to build a complete simulation of the human brain within a supercomputer.

Google X Lab has built an artificially intelligent robot – 2011-12-05 12:00:00

Someone claimed today that Google has developed a robot that can pass the Turing Test 93% of the time. The claim was posted anonymously to reddit.com by a supposed former Google X Lab employee. The posting is most likely a hoax, but interesting reading nevertheless. It possibly contains some elements of truth, although we don’t know which elements.

Silicon synapse built at MIT – 2011-11-19 12:00:00

Today researchers at MIT announced they have built a silicon synapse that models the ion channels in a single biological synapse. It contains ~400 transistors and operates using analog current, not digital. They plan to use the chip to investigate how biological synapses are strengthened and weakened, and to build larger systems that model neural functions such as the visual system. See the news article and the research paper.

Video presentation of the Human Brain Project – 2011-11-16 12:00:00

Newly released video presention of the Human Brain Project given by project founder Henry Markram on November 2, 2011 in San Francisco. Skip to chapter 8 of the video if you’re only interested in Markram’s part of the talk. He also talks with neuroscientist David Eagleman about whether a brain simulation would be truly self-aware.

Scientific papers describing IBM’s neurosynaptic core – 2011-10-07 12:00:00

Scientific papers describing IBM’s neurosynaptic core were published online yesterday. These cores implement 256 digital integrate-and-fire neurons and a 1024×256 bit memory for synapses. They use IBM’s 45nm SOI process. The chips are anticipated to become a key building block of a modular neuromorphic architecture.

Part two of the Blue Brain Project documentary video – 2011-08-22 12:00:00

Noah Hutton published part two of his video documentary about the Blue Brain Project today. It’s a nice clip and it’s great to get an update on the project’s progress. We meet the people involved and have a good look around their laboratory.

Scientists to simulate human brain inside a supercomputer

 

By Barry Neild, CNN updated 9:13 AM EDT, Fri October 12, 2012 |

Researchers say building a computer simulation could improve understanding and treatments of brain diseases like Alzheimer’s.

STORY HIGHLIGHTS

• Human Brain Project will use supercomputers to mimic tangle of neurons and synapses that power our thoughts
• Scientists say the simulator could offer new insight into the treatment of brain disease like Parkinson’s and Alzheimer’s
• “Brain in a box” is unlikely to transform into sci-fi-style computer bent on world domination, scientists say

(CNN) — There’s no escaping the fact that the Human Brain Project, with its billion-dollar plan to recreate the human mind inside a supercomputer, sounds like a science fiction nightmare.

But those involved hope their ambitious goal of simulating the tangle of neurons and synapses that power our thought processes could offer solutions to tackling conditions such as depression, Parkinson’s disease and Alzheimer’s.

The Human Brain venture is the next step in a long-running program that has already succeeded in using computers to create a virtual replica of part of a rat’s neocortex — a section of the brain believed to control higher functions such as conscious thought, movement and reasoning.

Scientists at its forerunner, the Switzerland-based Blue Brain Project, have been working since 2005 to feed a computer with vast quantities of data and algorithms produced from studying tiny slivers of rodent gray matter.

“This is a tool for research, not a giant simulated brain that is going to rule the world
Sean Hill, neuroscientist

Last month they announced a significant advancement when they were able to use their simulator to accurately predict the location of synapses in the neocortex, effectively mapping out the complex electrical brain circuitry through which thoughts travel.

Henry Markram, the South African-born neuroscientist who heads the project, said the breakthrough would have taken “decades, if not centuries” to chart using a real neocortex. He said it was proof their concept, dubbed “brain in a box” by Nature magazine, would work.

Read: Mapping out a new era in brain research

Now the team are joining forces with other scientists to create the Human Brain Project. As its name suggests, they aim to scale up their model to recreate an entire human brain.

It is a step that will need both a huge increase in funding and access to computers so advanced that they have yet to be built.

If their current bid for €1 billion ($1.3 billion) of European Commission funding over the next 10 years is successful, Markram predicts that his computer neuroscientists are a decade away from producing a synthetic mind that could, in theory, talk and interact in the same way humans do.

His bold claims have inevitably fueled comparisons to doom-laden popular fiction in which conscious machines turn on their creators and wreak havoc.

The project’s scientists have been referred to as “team Frankenstein” and their computer likened to “Skynet,” the virtual intelligence that unleashes a robot war on humanity in the “Terminator” films.

Sean Hill, a senior computational neuroscientist on the project, laughs at such comparisons.

He says the computer will primarily become a repository for knowledge about the brain that will allow scientists to conduct experiments without the need to probe inside people’s skulls.

Read: $1 billion mission to reach the Earth’s mantle

“This is a tool for research, not a giant simulated brain that is going to rule the world,” he said.

“Right now, we’re in a crisis in neuroscience. There’s a lot of wonderful data being gathered but we don’t have a place where we can put those experimental results together and understand their implications.

“We are just beginning to appreciate how complex our brains are, far beyond any other device in the known universe
Terry Sejnowski, Salk Institute for Biological Studies

“The benefit of having this facility is you have a place to integrate the data into a model where you can test predictions and start to learn principles of how the brain operates.”

The computing power needed to build the model is phenomenal. Simply to replicate one of the 10,000 neuron brain cells involved in the rat experiment took the processing capacity usually found in a single laptop. To simulate a fully functioning human brain, it would take billions.

Hill says that such computational power — known as exascale — will be available by the end of the decade. The Human Brain Project’s scientists are hoping to work with supercomputer developers to ensure future machines match their requirements.

But, even as the team touts its experiments as a possible solution to the brain diseases that affect about two billion people worldwide, they have attracted critics who say their work is far too broad in scope to achieve usable results.

Professor Terry Sejnowski, head of the Computational Neurobiology Laboratory at the Salk Institute for Biological Studies in San Diego, has been quoted as saying the Blue Brain project is “bound to fail.”

Read: How the search for aliens can help sustain life on Earth

He told CNN via email that “progress is being made but there is still a long way to go before we will understand the computational capabilities of cortical circuits.”

He added: “We are just beginning to appreciate how complex our brains are, far beyond any other device in the known universe.”

Sean Hill said the team hoped it was answering skeptics with its achievements so far.

“It’s just a matter of keeping on doing it. Let’s keep improving these tools and open them up so that many scientists are engaged and collaborating and using it as common point to bring the data together,” he said.

“The only way to address the critics is to keep working, showing the positive results and do the best we can — and that is starting to happen.”