Author Archives: Vinodh Ravindranath
Suppose you are asked to prove how useful a feature could be. Let's say you have to estimate the number of users on your platform that would benefit from a certain feature or change on your product. You pick a “random uniform sample” of 1% of the users. They are now exposed to the feature.…
The post Small Vs Random Samples: Understanding Underlying Probability appeared first on Analytics India Magazine.
A lot of us, particularly those who grew up in India, may have heard the Akbar — Birbal story, where the great Mughal emperor Akbar poses a seemingly impossible question to answer — “How many crows are there in our kingdom?”. Birbal, the witty and intelligent minister of Emperor Akbar, retorts with a random number…
The post A Data Science Question In The Times Of Akbar and Birbal appeared first on Analytics India Magazine.
Most machine learning problems boil down to fitting a set of observations(points) to some mathematical curve (simplistically referred to as the “model” in machine learning parlance). That yields a compressed representation that can be used for applications including prediction and pattern recognition. The process of fitting typically involves minimising what is called the cost function…
The post The World Of Cost Functions — Inclusivity, Majoritarianism And Oligarchy appeared first on Analytics India Magazine.
Most machine learning problems boil down to fitting a set of observations(points) to some mathematical curve (simplistically referred to as the “model” in machine learning parlance). That yields a compressed representation that can be used for applications including prediction and pattern recognition. The process of fitting typically involves minimising what is called the cost function…
The post The World Of Cost Functions — Inclusivity, Majoritarianism And Oligarchy appeared first on Analytics India Magazine.