Nowadays, technologies are continuously growing to reduce human efforts. Similarly, one more nanoscale device has been invented to power artificial systems that can mimic the human brain. Scientists from the University of Southampton have developed a new device called Memristor. The memristors are electrical components that limit or regulate the flow of electrical current in a circuit.
The memristor which acts like human brain could be used in pervasive sensing technologies. For example, fuel real-time monitoring in harsh or inaccessible environments. It offers a way towards that end by supporting many fundamental features of learning synapses in extremely compact volumes. If artificial brains are ever going to become reality, therefore, memristive synapses have to succeed.
Acting like synapses in the brain, the metal-oxide memristor array is capable of learning input patterns in an unsupervised manner within a probabilistic winner-take-all (WTA) network. Thus, it enables low power equipped processors.
Scientists showcased that a memristor known as Artificial neural networks (ANNs) displays learning abilities. According to scientists, it has a highly attractive capability to enable the Internet of Things vision. It even can perform tasks that are difficult for conventional computing systems. For example, pattern recognition, online learning, and classification.
Currently, practical ANNs are blocked due to the inefficiency of hardware synapses. Although, they are essential components, which every ANNs require in large quantity.
Now, scientists show that the memristor used inside the ANN, support for sophisticated learning rules. This is for carrying out reversible learning of noisy input data.
Lead author Dr. Alex Serb said, “If we want to build artificial systems that can mimic the brain in function and power we need to use hundreds of billions or trillions of artificial synapses. Many of which must be able to implement learning rules of varying degrees of complexity.”
“Although currently available electronic components would create such synapses. The required power and area efficiency benchmarks will be extremely difficult to meet without designing new and bespoke ‘synapse components’,” he added.
Themis Prodromakis said, “Our work establishes such a technological paradigm shift. It proves that nanoscale memristors will be used for formulating silicon neural circuits to process big data in real-time. It is a key challenge of modern society.”
“We have shown that such hardware platforms can independently adapt to its environment without any human intervention and are very resilient in processing even noisy data in real-time reliably,” he added.