What is Machine Learning?
Machine Learning is a sub-set of artificial intelligence where computer algorithms are used to autonomously learn from data and information. In machine learning computers don’t have to be explicitly programmed but can change and improve their algorithms by themselves.
When did this act of “Training Machines” begin?
A chronological arrangement of the crucial events in Machine Learning is recorded below:
1801 – Joseph Marie Jacquard, a French weaver and merchant, asserted that storing data was the next big challenge after the invention of Pascal’s Machine. The first use of storing data was in a weaving loom invented by Joseph Marie Jacquared that used metal cards punched with holes to position threads. A collection of these cards coded a program that directed the loom thus allowing for a process to be repeated with consistent result every time.
1950 — Alan Turing creates the “Turing Test” to determine if a computer has real intelligence. To pass the test, a computer must be able to fool a human into believing it is also human.
1952 — Arthur Samuel wrote the first Learning program. The program was the game of checkers, and the IBM computer improved at the game the more it played, studying which moves made up winning strategies and incorporating those moves into its program.
1957 — Frank Rosenblatt designed the first Neural Network for computers (the Perceptron), which can simulate the thought processes of the human brain.
1967 — The “Nearest Neighbor” algorithm was written, allowing computers to begin using very basic pattern recognition. This could be used to map a route for traveling salesmen, starting at a random city but ensuring they visit all cities during a short tour.
1981 — Development of Explanation Based Learning model.
Further, vigorous research into the field has given birth to countless Machine Learning models and Paradigms. Collectively, they have led to the development of systems that can predict stock prices and place such abilities in a machine, packing an engine inside its hood, to magically maneuver on its own.
Why do we need ML?
In present day’s Earth when every single human is fabricates data every moment it breaths, collectively producing tons of data, analyzing the underlying information in these data provides information that helps companies improve its services such as the instant product suggestions on Amazon, predict possible afflictions based on images and other details that cannot be processes, otherwise, in the blink of eye.
What is at the core of Machine Learning?
Actually, Machine Learning veils a bunch of complex mathematics involving lots of Probability(recall the process used to calculate the chances a coil will land to its ‘Heads’ or ‘Tails’) and Linear Algebra(remember Matrices and Determinants).
Why Linear Algebra and Probability?
So, why do we use these so extensively in ML?
Machine Learning, at its heart is making predictions through complex mathematical expressions that are strenous for humans to perform. For Linear Algebra, it turns out that the algorithms are laid out in a pattern that renders the use of matrices almost inevitable.
ML is further classified under different subsets based on how each model uses data to learn. Currently, they are split up under Supervised, Unsupervised and Reinforcement Learning.
Supervised Machine Learning: The program is “trained” on a pre-defined set of “training examples”, which then facilitates its ability to reach an accurate conclusion when given new data.
Unsupervised Machine Learning: The program is given a bunch of data and must find patterns and relationships therein.
Reinforcement Machine Learning: The crux of this model can be understood as a program that learns from its mistakes made while training, get rebooted from the point immediately prior to committing the mistake/error, and takes a new path to increase its efficiency.
Current advancements in the field: write about deep learning(tensor flow and how it’s different from the old)
From next week, the posts will delve deeper into the magical world of the process of ‘Teaching Machines’ and try to elucidate how ML does what it does.