Building Brain-Inspired AI Supercomputing System

IBM (NYSE: IBM) and the U.S. Air Force Research Laboratory (AFRL) today announced they are collaborating on a first-of-a-kind brain-inspired supercomputing system powered by a 64-chip array of the IBM TrueNorth Neurosynaptic System. The scalable platform IBM is building for AFRL will feature an end-to-end software ecosystem designed to enable deep neural-network learning and information discovery. The system’s advanced pattern recognition and sensory processing power will be the equivalent of 64 million neurons and 16 billion synapses, while the processor component will consume the energy equivalent of a dim light bulb – a mere 10 watts to power.
IBM researchers believe the brain-inspired, neural network design of TrueNorth will be far more efficient for pattern recognition and integrated sensory processing than systems powered by conventional chips. AFRL is investigating applications of the system in embedded, mobile, autonomous settings where, today, size, weight and power (SWaP) are key limiting factors. The IBM TrueNorth Neurosynaptic System can efficiently convert data (such as images, video, audio and text) from multiple, distributed sensors into symbols in real time. AFRL will combine this “right-brain perception capability of the system with the “left-brain” symbol processing capabilities of conventional computer systems. The large scale of the system will enable both “data parallelism” where multiple data sources can be run in parallel against the same neural network and “model parallelism” where independent neural networks form an ensemble that can be run in parallel on the same data.


AFRL was the earliest adopter of TrueNorth for converting data into decisions,” said Daniel S. Goddard, director, information directorate, U.S. Air Force Research Lab. “The new neurosynaptic system will be used to enable new computing capabilities important to AFRL’s mission to explore, prototype and demonstrate high-impact, game-changing technologies that enable the Air Force and the nation to maintain its superior technical advantage.”

“The evolution of the IBM TrueNorth Neurosynaptic System is a solid proof point in our quest to lead the industry in AI hardware innovation,” said Dharmendra S. Modha, IBM Fellow, chief scientist, brain-inspired computing, IBM Research – Almaden. “Over the last six years, IBM has expanded the number of neurons per system from 256 to more than 64 million – an 800 percent annual increase over six years.’’


Printing 3-D Graphene For Tissue Engineering

Ever since single-layer graphene burst onto the science scene in 2004, the possibilities for the promising material have seemed nearly endless. With its high electrical conductivity, ability to store energy, and ultra-strong and lightweight structure, graphene has potential for many applications in electronics, energy, the environment, and even medicine.

Now a team of Northwestern University researchers has found a way to print three-dimensional structures with graphene nanoflakes. The fast and efficient method could open up new opportunities for using graphene printed scaffolds regenerative engineering and other electronic or medical applications.
Led by Ramille Shah, assistant professor of materials science and engineering at the McCormick School of Engineering and of surgery in the Feinberg School of Medicine, and her postdoctoral fellow Adam Jakus, the team developed a novel graphene-based ink that can be used to print large, robust 3-D structures.

People have tried to print graphene before,” Shah said. “But it’s been a mostly polymer composite with graphene making up less than 20 percent of the volume.

With a volume so meager, those inks are unable to maintain many of graphene’s celebrated properties. But adding higher volumes of graphene flakes to the mix in these ink systems typically results in printed structures too brittle and fragile to manipulate. Shah’s ink is the best of both worlds. At 60-70 percent graphene, it preserves the material’s unique properties, including its electrical conductivity. And it’s flexible and robust enough to print robust macroscopic structures. The ink’s secret lies in its formulation: the graphene flakes are mixed with a biocompatible elastomer and quickly evaporating solvents

It’s a liquid ink,” Shah explained. “After the ink is extruded, one of the solvents in the system evaporates right away, causing the structure to solidify nearly instantly. The presence of the other solvents and the interaction with the specific polymer binder chosen also has a significant contribution to its resulting flexibility and properties. Because it holds its shape, we are able to build larger, well-defined objects.
An expert in biomaterials, Shah said 3-D printed graphene scaffolds could play a role in tissue engineering and regenerative medicine as well as in electronic devices. Her team populated one of the scaffolds with stem cells to surprising results. Not only did the cells survive, they divided, proliferated, and morphed into neuron-like cells.


Building an artificial brain

A scientific  team is creating a synapse using carbon nanotubes. Engineering researchers of  the University of Southern California have made a significant breakthrough in the use of nanotechnologies for the construction of a synthetic brain. They have built a carbon nanotube synapse circuit whose behavior in tests reproduces the function of a neuron, the building block of the brain. (Physorg April 21.2011).



Meanwhile a nanocomputer research team at Harvard University  have succeeded a major milestone towards the first nancocomputer, which if  connected to medical research can produce an huge progress for humanity. predicting the feasibility of artificial brains in the future. The researchers from USC focus on biomimetic neural models and electronic circuits that implement those models.  Complexities in modeling biological neural tissue are discussed.  Estimates are given for the size of artificial neural systems based on CMOS technology in 2021, without considering interconnections. Some solutions to the problem of interconnecting neurons are proposed.