Researchers have developed a brain like computing device that simulates human learning by connecting single synaptic transistors into a neuromorphic circuit. The research was published through the journal nature communications on 30th of April.
https://www.nature.com/articles/s41467-021-22680-5
Similar to how physiologist Pavlov conditioned
dogs to associate a bell with food, researchers at Northern University and
University of Hong Kong successfully conditioned their circuit to associate
light with pressure.
Ivan Pavlov's experiment
Ivan Pavlov researched on salivation in dogs in response to being fed. He inserted a test tube into the cheek of a dog to measure saliva when the dogs were being fed. He introduced a stimulus (eg. sound of a metronome), then gave the dog food, after a couple of repetitions, dog began to salivate in response to the stimulus. He called the dogs anticipatory salivation "psychic secretion".
Jonathan Rivney, is the senior author of this study. Rivney is an assistant professor of biomedical engineering at Northwestern's McCormick School of Engineering. With the assistance of Paddy Chan, associate professor of mechanical engineering at University of Hong Kong, he led the study. The paper's first author is Xudong Ji, a post doctoral researcher in Rivney's group.
The device consist of a totally unique organic electrochemical synaptic transistors which simultaneously process and store information a bit like the human brain. The researchers demonstrated that the transistor can mimic the short-term and long-term plasticity of synapses within the human brain building on memories to find out over time. Due to its brain-like ability the novel transistor (combines logic and memory functions, drastically reduces power consumption) could potentially overcome the limitations of traditional computing, including their energy-sapping hardware and the limited ability to perform multiple tasks at the same time. The brain-like device also has higher fault tolerance, continuing to work smoothly even when some components fail. Plasticity of the synapse is the basic building block of the brain's computational power. This is the reason why the human brain can outperform outstanding modern computers in some complex and unstructured tasks such as pattern recognition, motor control and multi sensory integration. These synapses enable the brain to function properly in a highly parallel fault tolerant and energy efficient manner. In their work, the researchers demonstrate an organic, plastic transistor that mimics key functions of a biological synapse.
Problem with conventional computing
Conventional digital computing systems have separate processing and storage units, causing data intensive tasks to consume large amounts of energy. Inspired by the combined computing and storage process within the human brain, researchers in recent years have sought to develop computers that operate more just like the human brain, with arrays of devices that function like a network of neurons. At present the computer system works in a way that its memory and logic are physically separated. Whenever we perform a computation it sends the information to a memory unit and every time to retrieve that information we need to recall it. If these separate functions can be brought together it is possible to save space and energy costs. Currently, the memory resistor or "memristor" (a non volatile Electronic memory device which is stable and remembers their state even if the device loses power is lost) is the most well developed technology which will perform combined processing and memory function. But memristors still suffer from energy costly switching and less bio compatibility. These problems led researchers to the synaptic transistor-especially the organic electrochemical synaptic transistor which operates with low voltages, continuously tuneable memory and high compatibility for biological applications. Still challenges exist .
Even high-performing organic electrochemical synaptic transistors require the write operation to be decoupled from the read operation.
- Rivnay
To overcome these challenges the North Western and therefore the University of Hong Kong team optimised a conductive, plastic material within the organic, electrochemical transistor which will trap ions.
How does Synaptic transistor work?
In the brain, a synapse is a structure through
which a neuron can transmit signals to a different neuron using small molecules
called neurotransmitters. In the synaptic transistor, ions behave similarly to
neurotransmitters, sending signals between terminals to make a man-made
synapse. By retaining stored data from trapped ions, the transistor remembers
previous activities, developing long-term plasticity. The researchers
demonstrated their device's synaptic behaviour by connecting single synaptic
transistors into a neuromorphic circuit to simulate associative learning. They
integrated light and pressure sensors in the circuit and trained the circuit to
associate both the unrelated physical inputs (pressure and light) with each
another.
Perhaps the foremost famous example of associative learning is Pavlov's dog which naturally drooled when it encountered food. After conditioning the dog to associate a bell ring with food, the dog also began drooling when it heard the sound of a bell. For the neuromorphic circuit, the researchers activated a voltage by applying pressure with a finger press. To condition the circuit to associate light with pressure, the researchers first applied pulsed light from an LED light bulb and then immediately applied pressure. In this scenario, the pressure represents the food and the light represents the bell. The device's corresponding sensors detected both inputs. After one training cycle, the circuit made an initial connection between light and pressure. After five training cycles, the circuit significantly associated light with pressure. Light, alone, was ready to trigger a signal or "unconditioned response".
Future applications
Since the synaptic circuit is made of soft polymers like plastic, it can be readily fabricated on flexible sheets and easily integrated into soft, wearable electronics, smart robotics and implantable devices that directly interfere with living tissue and even the brain. The researchers say that while their application is a proof of concept, their proposed circuit can be further extended to include more sensory inputs and integrated with other electronics to enable onsite, low power computation. Because it is compatible with biological environments the device can directly interfere with living tissue which is critical for next generation bioelectronics.