In many ways how human-kind has evolved the computer-systems is actually a reflection of their own self. If you want to compare very abstractly you could see that the state of affair of writing a computer application was very different before the Object Oriented Program (OOP) was manifested, but after OOP took over, the whole way we looked at computer programs was as though we are looking at our own self.

I remember, one of the very early examples that i had come across while learning OOP was imagining human being as a class. If you dissect this class more and look inside, you will find all sorts of things. The class’s behaviour and actions which were described in methods. There are various other classes which exist with a relationship with the human class. A hand class has its own functions , humans have hands and so there must be some kind of relationship between the human class and the hand class and we progressed in that direction and the object of human class can be as detailed as the architecture of that programmer has made it. The more details we get into of a particular class, (example a human has a hand which has fingers which has nails which is made of some molecules which are further made of the atoms and so on..), the more realistic our program and the more it will be indistinguishable from a real human arm.

Ok, Getting a little off-topic.

Of-course to understand something and how it functions its very essential to visualise it, If you could visualise something , there is a high chance that you understand it more deeply. 2 dimension is easier to visualise , 3 dimension is also understandable but only when its represented in its true form, anything more than 3 dimensions and we start having problems. A lot of time the meaning is lost when a 3 Dimensional approach or visualisation is shown in its 2 dimensional projection. Maybe because the resources are not available to compute it or the imagination cannot be expressed in its true form. Now wait, why are we talking about visualising 3D and 2D all of a sudden while talking about Classes and Objects.

Okay coming back to the OOP, maybe the human being classes and various human objects and their relationships can be described using OOP but it cannot even start to do justice when we might have to describe our brain. A brain on the other hand can be best described using a neural network which is a combination of many neurones. Though the very first artificial neurone was created back in 1943, it could not be used to do many many practical things because of our then computational limitations.

The visualisation of an artificial neural network today (at-least in all the scientific papers that we read) is done mostly using the 2D in your computer screen which can lead to not visualising it properly, when i need to visualise it while studying various neural networks which solve our current problems i imagine stars, at night, stars upon stars can help you visualise thousand of neurones. If i simply lookup into the sky on a clear night i can see layers upon layers and divided in various planes, array of stars, simply imagining stars in one plane as input layer, many adjoining planes layers (hidden layers) and another plane of stars as output layer. This let me imagine a neural network more vividly.

I suddenly find myself thinking that the artificial neural networks today and its many forms i.e CNN for Image recognition, RNN (LSTM) for language modelling takes up so much energy and the humans with their biological brains are able to do it so easily with much lesser power. But, yet we provide our CNNs & RNNs with the power it needs. There is a considerable amount of energy that goes in the development of these neural networks, (in form of powering up the data centre and cpu-clusters for the compute power) Why ? so that we can differentiate between photos and different breeds of dogs ?

Humans get all their energy from our sun which has led to the development of our biological neurones through evolution which keeps looking for an some sort of an ultimate answer. The reason why the sun is there , the reason why we all are here, the reason why our neurones have developed this way ? maybe one of the many models of our beloved AI through their many upon many epochs will one day give us those answers.

Till then, lets make models.