It does not take a keen mind to realize that the similarities that exist between the human brain and computer processing. This has been the case ever since the inception of computers and it is often posited by experts that the human brain is capable of operating at 1 exaFLOP, which comes down to a billion billion calculations per second. This is more than 5 times faster than that of Summit, the fastest supercomputer in the world that was built for the US Department of Energy’s Oak Ridge National Laboratory. It operates at a peak performance of 187.66 petaFLOP.
In essence, our brain is nothing more than a hotspot of electrical activity. It involves an array of interconnected areas that are weaved together by an intricate network of billions of neurons and trillions of glial cells. Eliminate the element of neuroplasticity (the ability of the brain to form new neural connections at will) from the equation and we have a fully functioning organ that is nothing more than a picture-perfect case of the most advanced supercomputer in the world.
Incidentally, this also begs the questions – what all similarities exist between the human body and computers? And if the right use case is identified, is it possible to reverse engineer the human functions to create new-age technologies that we can only dream of today?
To further this stance, let us dig into some examples and draw a comparison between the human body and computers.
Human Eyes and Neural Networks
If we go back to the drawing board and recall what we studied about the human eye in high school, we know that its functioning can be explained by the field of optics. But what you may have skipped is the fact that all the imagery is primarily driven by a network of millions of neurons that have each been hard-wired to observe only one kind of stimulus. This includes a network of about 4.6 million cones and 92 million rods. Further studies have also revealed that each of these neurons changes as the orientation of the stimuli is changed.
And this is exactly what was used as the foundation of convolutional neural networks. Neural networks today follow the concept of hierarchical processing, just like the human brain does. This enables two neurons that are respectively detecting vertical and horizontal lines to combine their observations to depict the presence of a curve.
Edge Computing and Processing
Edge Computing is based on the simple yet extravagant idea of processing data at the edge of networks, such as computers that are connected near IoT-enabled devices as compared to distant data centers. This helps applications to be processed and executed faster since the processors are stationed closer to the site of data collection. A great analogy to draw here is the functioning of traffic signals where processing (traffic management, time computation), stimuli (changing of traffic colors), and the ensuing action (stopping and movement of traffic) all occur at the site of processing itself.
Such a defined system is also helping experts to come up with more optimized computational models that are inspired by the human brain. This involves the development of low-powered Edge Computing architectures where process development and integration, circuit design, system architecture, and learning algorithms are optimized simultaneously.
Phagocytes and Continuous Development
Phagocytes are cells in the human body that protect the body from ingesting harmful foreign particles, bacteria, and dead or dying cells. This paves the way for the process of Phagocytosis where cells use their plasma membranes to engulf large particles, helping the body to get rid of pathogens and cell debris.
When studied and experimented upon controlled conditions in laboratories, it was revealed that Phagocytes do the job of capturing the bacterias that prevail in the body and either provide them extra oxygen or make them devoid of oxygen. The results are then saved in a repository of data so that the results can be used in the future.
In the practical sense, this is the exact process through which machine learning works. With the help of the right Machine Learning model, computers can understand the data transactions that are happening in the backend, differentiate good transactions from bad transactions, and learn about them, and replicate the behaviors in the future.
The human body, and particularly the human brain, has brought forth a paradigm shift in the manner in which technology is perceived and developed today. It is no longer a far-fetched version of siloed functions that exist like legacy technologies. Instead, such analogical innovations are giving rise to new-age tech like never seen before, clocking unceasing development with concepts such as spike coding and spike-timing-dependent-plasticity.
I hope this blog acts as a catalyst that breeds innovative thought processes and urges you to think outside the box. Looking forward to your views.