Research advances in brain-inspired computing

Artificial Intelligence or AI has been always associated with software. However, the researchers focused on improving AI and machine learning performance, not through software but hardware. They concentrated on neuromorphic computing and introduced hardware improvements by harnessing a quality known as "randomness".

The research by researchers in Han Wang's Emerging Nanoscale Materials and Device Lab at USC Ming Hsieh Department of Electrical and Computer Engineering and the Mork Family Department of Chemical has been published in the 'Nature Communications Journal'.

In the brain, randomness played an important role in human thought or computation. It is born from billions of neurons that spike in response to input stimuli and generate a lot of signals that may or may not be relevant. The decision-making process perhaps is the best-studied example of how our brain makes use of randomness. It allowed the brain to take a detour from past experiences and explore a new solution when making a decision, especially in a challenging and unpredictable situation.

"Neurons exhibit stochastic behavior, which can help certain computational functions," said a USC Ph.D. student Jiahui Ma and a lead author Xiaodong Yan (both equally contributed as first authors).

The team wanted to emulate neurons as much as possible and designed a circuit to solve combinatorial optimization problems, which are one of the most important tasks for computers to complete.

The thinking is that for computers to do this efficiently, they needed to behave more like the human brain (on super steroids) in terms of how they process stimuli and information, as well as make decisions.

In much simpler terms, we needed computers to converge on the best solution among all possibilities.

The researchers said, "The randomness introduced in the new device demonstrated in this work can prevent it from getting stuck at a not-so-viable solution, and instead continue to search until it finds a close-to-optimal result."

"This is particularly important for optimization problems," said corresponding author Professor Wang.

"If one can dynamically tune the randomness features, the machine for performing optimization can work more efficiently as we desire," Wang added.

The researchers achieved this dynamic "tuning" by creating a specialized device, a hetero-memristor. Unlike transistors which were logic switches inside a regular computer chip, the hetero-memristor combined memory and computation together. Memristors that had been developed prior normally had a two-terminal structure. The Viterbi team's innovation is in adding a third electrical terminal and modulating its voltage to activate the neuron-like device and to dynamically tune the stochastic features in its output, much like one heated up a pot of water and dynamically adjusted the temperature to control the activity of the water molecules, hence enabled the so-called simulated "cooling." This provided a level of control that earlier memristors did not have.

The researchers said, "This method emulates the stochastic properties of neuron activity."

In fact, neuron activity is perceived to be random but may follow a certain probability pattern. The hetero-memristors they developed introduced such probability-governed randomness into a neuromorphic computing circuit by the reconfigurable tuning of the device's intrinsic stochastic property.

This is thus a more sophisticated building block for creating computers that can tackle sophisticated optimization problems, which can potentially be more efficient. Also, they can consume less power.