What is Neuromorphic Computing?
Neuromorphic computing is a method of computer engineering in which elements of a computer are modelled after systems in the human brain and nervous system. The term refers to the design of both hardware and software computing elements.
Neuromorphic engineers draw from several disciplines including computer science, biology, mathematics, electronic engineering and physics to create artificial neural systems inspired by biological structures.
There are two overarching goals of neuromorphic computing.
- The first is to create a device that can learn, retain information and even make logical deductions the way a human brain can — a cognition machine.
- The second goal is to acquire new information — and perhaps prove a rational theory — about how the human brain works.
How does Neuromorphic Computing works?
The working of neuromorphic computing-enabled devices begins with the placement of Artificial Neural Networks (ANN) that comprise millions of artificial neurons. These neurons are similar to the human brain neurons.
Enabling a machine (computer) to act and work like the human brain, layers of these artificial neurons pass signals to one another. These electric signals or electric spikes convert input into an output that results in the working of neuromorphic computing machines.
The passing on of electric spikes or signals functions on the basis of Spiking Neural Networks (SNN). This spiking neural network architecture further enables an artificial machine to work like the human brain does and perform functions that humans can do on a daily basis.
This can involve visual recognition, interpretation of data, and a lot more such tasks. Since these artificial neurons only consume power when the electric spikes are passed through them, neuromorphic computing machines are low-power-consuming computers as compared to traditional computers.
By imitating the neuro-biological networks present in the human brain, neuromorphic computing machines work like a human brain and perform tasks efficiently and effectively.
Bringing on the ability to work like the human brain, neuromorphic computing has advanced the developments in the field of technology. The engineering of computers in the earlier times led to the generation of traditional computers that consumed a lot of space for functioning.
However, computers working on the basis of neuromorphic computing consume much less space with an in-built capability to work faster and better.
SNN (Spiking neural network):
Artificial neural networks that closely mimic natural neural networks are known as spiking neural networks (SNNs). In addition to neuronal and synaptic status, SNNs incorporate time into their working model. The idea is that neurons in the SNN do not transmit information at the end of each propagation cycle (as they do in traditional multi-layer perceptron networks), but only when a membrane potential – a neuron’s intrinsic quality related to its membrane electrical charge – reaches a certain value, known as the threshold.
The neuron fires when the membrane potential hits the threshold, sending a signal to neighbouring neurons, which increase or decrease their potentials in response to the signal. A spiking neuron model is a neuron model that fires at the moment of threshold crossing.
Artificial neurons, despite their striking resemblance to biological neurons, do not behave in the same way. Biological and artificial NNs differ fundamentally in the following ways:
- Structure in general
- Computations in the brain
- In comparison to the brain, learning is a rule.
Alan Hodgkin and Andrew Huxley created the first scientific model of a Spiking Neural Network in 1952. The model characterized the initialization and propagation of action potentials in biological neurons. Biological neurons, on the other hand, do not transfer impulses directly. In order to communicate, chemicals called neurotransmitters must be exchanged in the synaptic gap.SNNs lag ANNs in terms of accuracy, but the gap is decreasing, and has vanished on some tasks.
Features of Neuromorphic Computing:
- Rapid Response System
- Low Consumption of Power
- High Adaptability
- Fast-paced Learning
- Mobile Architecture
Future of Neuromorphic Computing:
In simple terms, Artificial Intelligence future is Neuromorphic Computing. Setting forth the third wave or era of AI, neuromorphic computing will take over the technological advancements of the field and become the driving force of artificial intelligence future scope.
While the current wave of AI is faced with a number of challenges like heavy processing hardware and software storage capacity, the third wave of neuromorphic computing in AI will most likely put a stop to these challenges and enable the human-like activities performed by computers.
Neuromorphic chips, being manufactured by big tech giants like IBM, will be the key factor in making computers function like the human nervous system.
IBM Truenorth Chip:
TrueNorth was a neuromorphic CMOS integrated circuit produced by IBM in 2014. It is a manycore processor network on a chip design, with 4096 cores, each one having 256 programmable simulated neurons for a total of just over a million neurons. In turn, each neuron has 256 programmable “synapses” that convey the signals between them. Hence, the total number of programmable synapses is just over 268 million (228). Its basic transistor count is 5.4 billion.
According to IBM, it doesn’t have a clock and operates on unary numbers and computes by counting up to a maximum of 19 bits.The said cores are event-driven by using both synchronous and asynchronous logic and are interconnected through an asynchronous packet-switched mesh network on chip (NOC).IBM has developed a whole new ecosystem to program and use TrueNorth. It included simulator, a new programming language, an integrated programming environment and even libraries. This lack of backward compatibility with any previous technology (e.g. C++ compilers) poses serious vendor lock-in risks and other adverse consequences that may prevent it from commercialization in the future.
Applications of Neuromorphic Computing:
- Large Scale Operations: Elements of large-scale projects and product customization could also benefit from the use of neuromorphic computing. It could be used to more easily process large sets of data from environmental sensors. These sensors could measure water content, temperature, radiation, and other parametersdepending on the needs of the industry. The neuromorphic computing structure could help recognize patterns in these data, making it easier to reach effective conclusions.
- Artificial Intelligence: By definition, the field of neuromorphic computing strives to emulate the functionality of the human brain. The way the brain’s neurons receive, process, and send signals is extremely fast and energy-efficient. As such, it is natural that professionals in technology, specifically those in the field ofAI would be intrigued by this type of computing. As the name suggests, individuals in the field of AI focus on a particular element of brain intelligence. Intelligence is the ability of the mind to collect and apply information. Since this concept relates so closely to neuromorphic computing, it would be beneficial for the two fields to collaborate going forward.
– By Anirrvhinya, Third Year Department of