Skip to main content

Why Study the Brain?

“There is no scientific study more vital to man than the study of his own brain. Our entire view of the universe depends on it.” ― Francis Crick





After unraveling the mysteries of DNA, the secret to life, Francis Crick for the rest of his life turned his attention to solve the mysteries of brain and consciousness. He was certain that the answer to intelligence lies deep in the structure of brain. In his book, The Astonishing Hypothesis: The Scientific Search for the Soul, Francis Crick theorizes a framework for studying consciousness. His work turned out to be an inspiration for many AI researchers, as it became evident that deciphering brain might lead to creating general intelligence. Today the research from neuroscience is used by AI researchers to create intelligent algorithms, that are different than traditional symbolic based systems. We at Cere Labs try to draw inspiration from biology and other sciences to get insights in the research we conduct.


The discovery of Santiago Ramón y Cajal
The neuron is the structural and functional unit of the nervous system - Santiago Ramón y Cajal

Ramon y Cajal in his laboratory.

The path breaking discovery of Santiago Ramon Cajal that the neocortex is made up of many layers and that the neuron is the structural and functional unit of nervous system helped AI researchers to take inspiration from this model. The drawings made by Santiago were  precise and helped neuroscience to grow with an alarming rate. This is the reason he is called the father of modern neuroscience.

Neocortex is uniform
“The neocortex is uniform” - Vernon Benjamin Mountcastle

Vernon Benjamin Mountcastle discovered that the neocortex is uniform throughout. He determined that the brain, unlike any other part of the human body, is divided into little subunits, each with its own specific role.

This discovery led to the assumption that the neocortex uses the same mechanism for any kind of sensor including sensors for light, sound and touch. Take your brain and give it any sensor including sensors such as radar, the brain will eventually learn to understand data coming from the sensors.

Such discoveries made in neuroscience helps AI researchers conclude that pattern recognition is the key to Artificial General Intelligence. Today Deep Learning works on the principles of pattern recognition, and takes numerous inputs from many years of brain research.


Comments

Popular posts from this blog

Implement XOR in Tensorflow

XOR is considered as the 'Hello World' of Neural Networks. It seems like the best problem to try your first TensorFlow program.

Tensorflow makes it easy to build a neural network with few tweaks. All you have to do is make a graph and you have a neural network that learns the XOR function.

Why XOR? Well, XOR is the reason why backpropogation was invented in the first place. A single layer perceptron although quite successful in learning the AND and OR functions, can't learn XOR (Table 1) as it is just a linear classifier, and XOR is a linearly inseparable pattern (Figure 1). Thus the single layer perceptron goes into a panic mode while learning XOR – it can't just do that. 

Deep Propogation algorithm comes for the rescue. It learns an XOR by adding two lines L1 and L2 (Figure 2). This post assumes you know how the backpropogation algorithm works.



Following are the steps to implement the neural network in Figure 3 for XOR in Tensorflow:
1. Import necessary libraries
impo…

Understanding Generative Adversarial Networks - Part II

In "Understanding Generative Adversarial Networks - Part I" you gained a conceptual understanding of how GAN works. In this post let us get a mathematical understanding of GANs.
The loss functions can be designed most easily using the idea of zero-sum games. 
The sum of the costs of all players is 0. This is the Minimax algorithm for GANs
Let’s break it down.
Some terminology: V(D, G) : The value function for a minimax game E(X) : Expectation of a random variable X, also equal to its average value D(x) : The discriminator output for an input x from real data, represents probability G(z): The generator's output when its given z from the noise distribution D(G(z)): Combining the above, this represents the output of the discriminator when 
given a generated image G(z) as input
Now, as explained above, the discriminator is the maximizer and hence it tries to 
maximize