IBM and Lawrence Livermore National Laboratory have launched a new project to build a brain-inspired supercomputing platform for deep learning inference.
The project will be built on IBM’s TrueNorth chip, which the company claims will process the equivalent of 16 million neurons and 4 billion synapses and consume the energy equivalent of a tablet computer. The neural network design of IBM’s Neuromorphic System aims to be able to infer complex cognitive tasks such as pattern recognition and integrated sensory processing in a much more economical manner than current chips.
“The delivery of this advanced computing platform represents a major milestone as we enter the next era of cognitive computing,” said Dharmendra Modha, Chief Scientist for Brain-inspired Computing at IBM Research. “We value our relationships with the national labs. In fact, prior to design and fabrication, we simulated the IBM TrueNorth processor using LLNL’s Sequoia supercomputer. This collaboration will push the boundaries of brain-inspired computing to enable future systems that deliver unprecedented capability and throughput, while helping to minimize the capital, operating and programming costs – keeping our nation at the leading edge of science and technology.”
The technology itself will be utilized in a number of different manners within the National Nuclear Security Administration (NNSA), including the organizations Stockpile Stewardship Program, a program of reliability testing and maintenance of its nuclear weapons without the use of nuclear testing.
“Neuromorphic computing opens very exciting new possibilities and is consistent with what we see as the future of the high performance computing and simulation at the heart of our national security missions,” said Jim Brase, Livermore National Laboratory’s Deputy Associate Director for Data Science. “The potential capabilities neuromorphic computing represents and the machine intelligence that these will enable will change how we do science.”
While Artificial Intelligence has been one of the more prominent trends in the cloud computing world, the success of the technology and PR stunts launched has been varied.
AlphaGo is an example of the success of AI, as Google Deepmind’s AI program beat world Go champion Lee Se-dol in a five match series. As traditional machine learning techniques could not be applied in this instance, the team combined an advanced tree search with deep neural network allowing the program to readjust its behaviour through reinforcement learning. The win came as a surprise to commentators, as the game Go relies on intuition and feel.
On the opposite end of the spectrum, Microsoft has had to release an apology after its twitter inspired AI stunt backfired. The program tweeted controversial comments as it was unable to grasp the politically incorrect nature of the messages it received from users, as reported by the Independent.