- What Does a Lady Tasting Tea Have to Do with Science? - May 31, 2019.
Design of Experiments (DOE) is a statistical concept used to find the cause-and-effect relationships. Surprisingly, an experiment arising from a casual conversation about tea-drinking is one of the first examples of an experiment designed using statistical ideas.
Design of Experiments, Randomization, Statistics
- Are Vectorized Random Number Generators Actually Useful? - Aug 28, 2018.
I reported that you can multiply the speed of common (fast) random number generators such as PCG and xorshift128+ by a factor of three or four by vectorizing them using SIMD instructions. Is this actually useful in practice?
Parallelism, Programming, Random, Randomization
- The Surprising Complexity of Randomness - Jun 15, 2017.
The reason we have pseudorandom numbers is because generating true random numbers using a computer is difficult. Computers, by design, are excellent at taking a set of instructions and carrying them out in the exact same way, every single time.
Complexity, Probability, Random, Randomization
- Embrace the Random: A Case for Randomizing Acceptance of Borderline Papers - May 16, 2016.
A case for using randomization in the selection of borderline academic papers, a particular use case which has parallels with many other possible scenarios.
Academics, ICML, NIPS, Random, Randomization
- 3 Ways to Test the Accuracy of Your Predictive Models - Feb 8, 2014.
3 different methods for testing accuracy of predictive models from 3 leading analytics experts - Karl Rexer, John Elder, and Dean Abbott explain using lift charts, randomization testing, and bootstrap sampling.
Bootstrap sampling, Dean Abbott, Decile tables, John Elder, Karl Rexer, Lift charts, Predictive Models, Randomization, Target shuffling