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.

CLICK ON THE IMAGE TO ENJOY THE VIDEO

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.’’

Source: https://www-03.ibm.com/

NanoComputers That Imitate Human Brain

Making a nanocomputer that learns and remembers like a human brain is a daunting challenge. The complex organ has 86 billion neurons and trillions of connections — or synapses — that can grow stronger or weaker over time. But now scientists from the Tsinghua University (China) report in ACS’ journal Nano Letters the development of a first-of-its-kind synthetic synapse that mimics the plasticity of the real thing, bringing us one step closer to human-like artificial intelligence.

brain_synapse_hero

While the brain still holds many secrets, one thing we do know is that the flexibility, or plasticity, of neuronal synapses is a critical feature. In the synapse, many factors, including how many signaling molecules get released and the timing of release, can change. This mutability allows neurons to encode memories, learn and heal themselves. In recent years, researchers have been building artificial neurons and synapses with some success but without the flexibility needed for learning. Tian-Ling Ren and colleagues set out to address that challenge.

The researchers created an artificial synapse out of aluminum oxide and twisted bilayer graphene. By applying different electric voltages to the system, they found they could control the reaction intensity of the receiving “neuron.” The team says their novel dynamic system could aid in the development of biology-inspired electronics capable of learning and self-healing.

Source: http://pubs.acs.org/

How Bipolar Disorders Affect The Brain

A nano-sized discovery by Northwestern Medicine® scientists helps explain how bipolar disorder affects the brain and could one day lead to new drug therapies to treat the mental illness.

Scientists used a new super-resolution imaging method — the same method recognized with the 2014 Nobel Prize in chemistry — to peer deep into brain tissue from mice with bipolar-like behaviors. In the synapses (where communication between brain cells occurs), they discovered tiny “nanodomain” structures with concentrated levels of ANK3 — the gene most strongly associated with bipolar disorder risk. ANK3 is coding for the protein ankyrin-G.

synapse

We knew that ankyrin-G played an important role in bipolar disease, but we didn’t know how,” said Northwestern Medicine scientist Peter Penzes, corresponding author of the paper. “Through this imaging method we found the gene formed in nanodomain structures in the synapses, and we determined that these structures control or regulate the behavior of synapses.”

High-profile cases, including actress Catherine Zeta-Jones and politician Jesse Jackson, Jr., have brought attention to bipolar disorder. The illness causes unusual shifts in mood, energy, activity levels and the ability to carry out day-to-day tasks. About 3 percent of Americans experience bipolar disorder symptoms, and there is no cure. Recent large-scale human genetic studies have shown that genes can contribute to disease risk along with stress and other environmental factors. However, how these risk genes affect the brain is not known.

Penzes is a professor in physiology and psychiatry and behavioral sciences at Northwestern University Feinberg School of Medicine. The results were published Oct. 22 in the journal Neuron.

source:  http://www.northwestern.edu/