Researchers create a high-performance, low-energy artificial synapse for neural network computing
A new organic artificial synapse devised at Stanford university could support computers that better recreate the processing that occurs in the human brain. It could also lead to improvements in brain-machine technologies. These results are published in Nature Materials, and Yoeri van de Burgt of TU/e is one of the leading authors that developed this device.
Scientist aspire to create devices that match the human brain’s ability to learn and remember information while using little energy. Brainier computers could boost the capabilities of many up and coming technologies, including visual-based search, voice-controlled interfaces and driverless cars. Researchers have already built high-performance neural networks supported by artificially intelligent algorithms but these are still distant imitators of the brain. They are especially hindered by their dependence on the traditional computer hardware, which requires large amounts of power in order to store information.
Now, in research published in the Feb. 20 issue of Nature Materials, researchers at Stanford University in collaboration with Sandia National Laboratories have created an artificial synapse – a synthetic version of the space over which neurons in the brain send messages. Like a real synapse, they can train their device to hold specific electrical states that allow processing and memory to occur together. In simulations, their device allowed for highly accurate neural network performance while maintaining low energy use.
“It works like a real synapse but it's an organic, electronic device that can be engineered,” said Alberto Salleo, associate professor of materials science and engineering at Stanford and senior author of the paper. “It's an entirely new family of devices because this type of architecture has not been shown before. For many key metrics, it also performs better than anything that's been done before with inorganics.”
Building a brain
When we learn, electrical signals are sent between neurons in our brain. The most energy is needed the first time a synapse is traversed. Every time afterward, the connection requires less energy. This is how synapses efficiently facilitate both learning something new and remembering what we’ve learned. The artificial synapse, unlike most other versions of brain-like computing, also fulfills these two tasks simultaneously, and does so with substantial energy savings.
“Deep learning algorithms are very powerful but they rely on a processors to calculate and simulate the electrical states and store it somewhere else, which is inefficient in terms of energy and time,” said Yoeri van de Burgt, former postdoctoral scholar in the Salleo lab and lead author of the paper. “Instead of simulating a neural network, our work is trying to make a neural network.”
Van de Burgt started working as assistant professor at TU/e at the end of last year. He intends to continue the research in Eindhoven, in the Microsystems group, building an actual array of artificial synapses for further research and testing.
The artificial synapse is based off a battery design. It consists of two thin, flexible films with three terminals, connected by an electrolyte of salty water. The device works as a transistor, with one of the terminals controlling the flow of electricity between the other two.
Like a neural path in a brain being reinforced through learning, the researchers program the artificial synapse by discharging and recharging it repeatedly. Through this training, they have been able to predict within 1 percent of uncertainly what voltage will be required to get the synapse to a specific electrical state and, once there, it remains at that state. In other words, unlike a common computer, where you save your work to the hard drive before you turn it off, the artificial synapse can recall its programming without any additional actions or parts.
Testing a network of artificial synapses
Only one artificial synapse has been produced but researchers at Sandia National Laboratory used 15,000 measurements from experiments on that synapse to simulate how an array of them would work in a neural network. They tested the simulated network’s ability to recognize handwriting of digits 0 through 9. Tested on three datasets, the simulated array was able to identify the handwritten digits with an accuracy between 93 to 97 percent.
Although this task would be relatively simple for a person, traditional computers have a difficult time interpreting visual and auditory signals.
“More and more, the kinds of tasks that we expect our computing devices to do require computing that mimics the brain because using traditional computing to perform these tasks is becoming really power hungry,” said A. Alec Talin, distinguished member of technical staff at Sandia National Laboratories in Livermore, California and the other senior author of the paper. “What we've done is we've demonstrated a device that’s ideal for running these type of algorithms and consumes a lot less power.”
Whereas digital transistors can only be in two states, such a 0 and 1, the researchers showed they could successfully program 500 states in artificial synapse. In switching from one state to another they only used 10 picojoules of energy . They only tested the synapse in large devices and believe they could attain neuron-level energy efficiency once it’s tested in smaller devices. These two features make the device extremely well-suited for the kind of signal identification and classification that traditional computers struggle to perform.
Every part of the device is made of inexpensive, organic materials. These aren’t found in nature but they are largely comprised of hydrogen and carbon and are compatible with the brain’s chemistry. Cells have been grown on these materials and they have even been used to make artificial pumps for neural transmitters. The voltages applied to train the artificial synapse are also the same as those that move through human neurons.
All this means it’s possible that the artificial synapse could communicate with live neurons, leading to improved brain-machine interfaces. The softness and flexibility of the device also lends itself to being used in biological environments.