Most Winning A/B Test Results are Illusory
Whitepaper about errors in A/B testing, written for Qubit.
Denoising Dirty Documents – Part 10
In my last blog, I explained how to take advantage of an information leakage regarding the repeated backgrounds in Kaggle’s Denoising Dirty Documents competition. The result of that process was that we had done a fairly good job of removing the background. But the score from doing this was not good enough to get a good placing. We need to do some processing on the image to improve the score.
History of Monte Carlo Methods - Part 2
This is the second part of a three part post. The last part covered the early history of Monte Carlo and the rejection sampling method.
Q-learning with Neural Networks
What is Q-learning?¶
Spying on instance methods with Python's mock module
** Thu 29 October 2015
Hogwild Stochastic Gradient Descent
How SGD works
The Evolution of Pop Lyrics and a tale of two LDA’s
Go easy on Volkswagen
 |
I don’t want to imply that what VW did was acceptable. The act by person(s) there of deliberately programing engine management systems to cheat at emissions tests crossed far over the line from being an ‘optimization’ and pushed it squarely into the territory of misrepresentation. Call it what you will: distortion, falsification, deception, perversion, manipulation, cheating … it was a deliberate and conscious programming act. Someone thought about it, then implemented it. It was done on purpose. It was wrong. |
Books for Data Science Beginners, and Data Sources
I just wanted to note here on Becoming A Data Scientist that I recently wrote two posts over on Data Sci Guide that are getting some attention