On synapse intelligence emulated in self-formed artificial synaptic networks

In recent years, the technology is taking a big leap towards artificial intelligence. Software-based artificial neural network (ANN) is well established now and can be seen defeating humans in games (AlphaGo and AlphaZero) or helping to handle the Covid-19 situation. However, the power-hungry von Neumann architecture slows down ANNs performance due to available serial processing while, brain does the job parallelly with just 20 W. To overcome this bottleneck, a hardware-based solution is, realizing an artificial synaptic device which, unlike transistors, should be able to emulate the functions of human brain synapse. Current literature provides highly involved fabrication processes to realize a single synaptic junction only to emulate basic brain activity with the aid of CMOS electronics. This challenge has been well addressed by the nominee’s group following a novel approach of fabricating an artificial synaptic network (ASN) resembling the biological neural network via a self-forming method. The metal islands and nanoparticles in ASN resemble bio-neurons and neurotransmitters. This hierarchical structure facilitates various learning activities such as STP, LTP, potentiation, depression, associative learning, interest-based learning, supervision and impression of supervision. Remarkably, all these behaviors are emulated in a single material system without the aid of external CMOS circuits. A prototype kit developed to emulate the famous Pavlov’s dog behavior clearly demonstrates the potential of the device towards advanced neuromorphic artificial intelligence. This is an interesting case where chemical methods have been employed to realise a synaptic device!