You’re reading this post because an algorithm brought it to you among others to click, and you did, and the algorithm took note. Given this, you’d want to know how these algorithms work, constantly shaping the world around us. Before the advent of artificial intelligence, humans programmed machines with a set of logical instructions that humans could understand, and explain. However, many problems are too complex for humans to write instructions for, such as image recognition. This is where machine learning comes into the picture!
So, what is Machine Learning (ML)? Even among industry-level practitioners, there is no consensus regarding a go-to definition for machine learning. Let’s try to unpack one of the oldest definitions, and understand what this buzzword really means!
In 1959, Arthur Samuel, a pioneer in the field of machine learning defined it as – “the field of study that gives computers the ability to learn without being explicitly programmed”. This definition seems to be motivated by the Samuel Checkers Playing Program, which played games against itself and learned from them over time. Since a computer has the patience to play millions of games, unlike most humans, the program delivered remarkable results – ultimately (of course!) becoming a better Checkers player than Samuel himself. This program was among the world’s first successful self-learning programs, and as such a very early demonstration of artificial intelligence.
Machine learning really boils down to a way for computers to learn from experience – to be able to better generalize, and make promising predictions in the future.
Read more:
1. Arthur Samuel’s Checkers Playing Program
2. More about Machine Learning (MIT Technology Review)
- The Seven Bridges of Königsberg - August 17, 2020
- The Turing Test - August 6, 2020
- What is Machine Learning? - July 28, 2020