By mimicking how the brain operates, neuromorphic computing can use dramatically less energy than conventional electronic AI chips. However, even the most sophisticated neuromorphic devices today are still quite simple, using only a small fraction of the number of connections found in human neurons.
Now a new study suggests that using sound waves, neuromorphic devices can better mimic biological neurons and operate faster and with greater energy efficiecy than their electronic counterparts.
“This could make future neuromorphic hardware more compact, more parallel, and more efficient for tasks that require combining many features, such as pattern recognition, sensory processing, and data analysis,” says Xiaodong Yan, an assistant professor of materials science and engineering and electrical and computer engineering at the University of Arizona in Tucson.
Just as brains use synapses—the links connecting neurons—to help them both compute and store data, neuromorphic devices often combine both operations. Doing so can reduce the energy and time needed for conventional microchips to shuttle data between processors and memory.
Each human neuron may have thousands of synapses connecting them with other cells; one kind of neuron found in the cerebellum, the Purkinje cell, may have as many as 100,000 synapses. This extraordinary level of connectivity lets each human neuron “combine different pieces of information, compare them, and respond depending on the context,” Yan says.
In contrast, most conventional neuromorphic devices are essentially “one artificial synapse,” Yan says. Building an artificial neuron with as many synapses as a human neuron would require wiring many separate devices together. “This increases wiring, energy cost, and hardware complexity,” Yan says.
Recently, scientists have developed acoustic devices in which sound waves can encode multiple values in its waves’ phase. These phase bits, or phi-bits, can in turn support quantum-like logic gates and parallel computing. Whereas conventional bits each only symbolize two digits, 0 or 1, and require a separate physical component for each bit, phi-bits each represent multiple variables and coexist within one space.
To be clear, however, operations on phi-bits are not quantum computations, only classical analogues of quantum computer systems.
Now Yan and his colleagues have developed an acoustic synapse containing multiple phi-bits. This enables multiple simultaneous computations in a relatively simple way, with lower power requirements compared to conventional electronics.
“The idea of bringing new physics to more efficiently perform complex computations is always fascinating,” says Brad Aimone, a researcher at the Center for Computing Research at Sandia National Laboratories in Albuquerque, N.M..
“It opens new opportunities worth thinking about, going forward,” says Aimone, who did not take part in this study.
The new device the scientists developed consists of three aluminum rods, each roughly 60 centimeters long and 1.25 centimeters wide, and connected by epoxy glue. The researchers used a thin layer of honey to attach ultrasonic transmitters and sensors to the ends of the rods.
Yan and his colleagues used sound waves to encode a stream of data, including images and labels that identified those images. The ultrasonic transmitters emitted these sound waves through the rods, which interact acoustically via the epoxy. Ultrasonic sensors in the device then detected the acoustic signals from the acoustic interactions.
The researchers found they could modulate the phase of phi-bits in ways that mimicked the ability of biological synapses to strengthen or weaken over time, part of why memories last or fade. This property, called synaptic plasticity, meant the researchers could train their acoustic synapse to perform a range of tasks.
In experiments, the scientists tested a topological acoustic synapse coupled with three digital neurons. (The emerging field of topological acoustics, applying previously unknown properties of sound waves, has led to new ways to manipulate sound—for instance, in circuits in which sound waves can flow with virtually no dissipation of energy.) “In a topological acoustic synapse, the acoustic wave interactions help transform and organize information before the final readout,” Yan says.
When it came to classifying 150 flowers as belonging to one of three iris species, the new device outperformed a conventional computer chip-based neural network called a multilayer perceptron (MLP). The acoustic device—representing a single simulated synapse—achieved a final accuracy of 96.7 percent using only 39 parameters and reached its peak accuracy 20 percent faster than MLPs. To achieve comparable accuracy, the researchers note an MLP would require nine neurons and even more parameters.
All in all, the researchers estimated their new device consumes at most one-tenth the power of current state-of-the-art electronic neuromorphic hardware. “Future neuromorphic systems may combine physical wave dynamics with conventional computing to achieve more energy-efficient information processing,” Yan says.
In addition, the scientists noted their new device could mimic the activity of critical molecules known as neuromodulators. Neuromodulators such as dopamine or serotonin “can make a synapse more sensitive, less sensitive, faster, slower, or change how strongly it learns,” Yan says. “This flexibility helps the brain adapt to different conditions, such as attention, reward, stress, or learning state.”
A single biological synapse may be simultaneously influenced by as many as 10 neuromodulators. However, mimicking neuromodulation in conventional neuromorphic hardware is challenging, typically requiring dramatically more complex designs.
Yet the researchers found that, with an acoustic synapse, simply adding an extra rod allowed the system to mimic a number of neuromodulatory processes—including rapid responses (such as dopamine effects on synaptic strength during learning) and slow, long-term responses (such as chronic stress).
“Neuromodulators let the brain use one circuit to perform different functions depending on the context,” Aimone says. “This is unlike now, where you have to have different neural networks for different tasks. So instead of an enormous neural network, you could have smaller neural networks that can use the equivalent of neuromodulators to adjust themselves for whatever’s going on. That’s really exciting.”
The researchers published their findings online 12 June in the journal * Science Advances*.