BM has brought artificial intelligence (AI) one step closer to reality by creating technology that imitates the brain’s neurons. The company’s scientists have created randomly spiking neurons using phase-change materials to store and process data, which is a “significant step forward in the development of energy-efficient, ultra-dense integrated neuromorphic technologies for applications in cognitive computing,” according to the company.
These artificial neurons are able to mimic the human brain in the way that they store and process data, and like an actual brain, they use little energy. Until now, achieving this has been a significant challenge for scientists.
“We have been researching phase-change materials for memory applications for over a decade, and our progress in the last 24 months has been remarkable,” said IBM Fellow Evangelos Eleftheriou.
“In this period, we have discovered and published new memory techniques, including projected memory, stored three bits per cell in phase-change memory for the first time, and now are demonstrating the powerful capabilities of phase-change-based artificial neurons, which can perform various computational primitives such as data-correlation detection and unsupervised learning at high speeds using very little energy.”
According to the scientists, the artificial neurons consist of phase-change materials, including geranium antimony telluride, which exhibit two stable states, an amorphous one (without a clearly defined structure) and a crystalline one (with structure). These materials are the basis of re-writable Blu-ray discs, however the artificial neurons do not store digital information; they are analog, just like the synapses and neurons in our biological brain.
In a published demonstration, the scientists applied a series of electrical pulses to the artificial neurons, which led to the progressive crystallisation of the phase-change material, causing the neuron to fire. In neuroscience, this function is known as the integrate-and-fire property of biological neurons. This is the foundation for event-based computation and is similar to how our brain triggers a response when we touch something hot.
Exploiting this integrate-and-fire property, even a single neuron can be used to detect patterns and discover correlations in real-time streams of event-based data, according to the scientists. For example, the artificial neurons could be used to detect patterns in financial transactions to find discrepancies or use data from social media to discover new cultural trends in real time. Large populations of these high-speed, low-energy nano-scale neurons could also be used in neuromorphic coprocessors with co-located memory and processing units.
The scientists have organised hundreds of artificial neurons into populations and used them to represent fast and complex signals. Moreover, the artificial neurons have been shown to sustain billions of switching cycles, which would correspond to multiple years of operation at an update frequency of 100Hz. The energy required for each neuron update was less than five picojoules and the average power less than 120 microwatts – for comparison, 60 million microwatts are needed to power a 60 watt lightbulb.