- Machine Learning and Deep Learning Explained
- Neural Networks and Natural Language Processing
- AI Algorithms: From Basics to Advanced
Machine Learning and Deep Learning Explained
Machine Learning is a field of study that aims to build machines "learning" from data.
In 1957, the scientist Frank Rosenblatt published (The Perceptron: A Perceiving and Recognizing Automaton, Cornell Aeronautical Laboratory) the first concept of perceptron, which is the learning algorithm based on human neuron behavior.
Rosenblatt's Perceptron study was based on the results of two other scientists (Pitts W. and McCulloch W., 1943).
Below, an image that depicts the human neuron:
The McCulloch-Pitts neuron works like a logic gate, where:
- The dendrites are the inputs
- The axon is the output that can be linked to other neurons
- The nucleus of the cell receives all the inputs
Furthermore, if the accumulated signal, in the nucleus, exceeds a threshold an output signal is generated and that will be passed on by the axon.
A logic schema below:
Where:
- x1, x2 … xn are the input values
- w1, w2 … wn are the weights on each inputs
- sum = x1 * w1 + x2 * w2 + … + xn * wn
- if sum > threshold output is 1 (or true) else it is 0 or (false)