Drawing From a Random Distribution in SQL

There are different methods for updating a reinforcement learning agent’s policy at each iteration. A few weeks ago we started experimenting with replacing our current method with a Bayesian inference step. Some of the data workloads within our agent are written in SQL that is executed on GCP’s BigQuery engine. We use this stack because it provides scalable computational capabilities, ML packages and a straightforward SQL interface. The Bayesian inference step we looked to implement uses one of the <a href="https://en.wikipedia.org/wiki/Beta_distribution#Mean_and_sample_size" rel="noopener ugc nofollow" target="_blank">alternative parametrizations of a beta distribution</a>. Which means that we will need to be able to draw from a beta distribution in SQL. While working on this, I discovered that drawing from the random distribution in SQL is a topic with very few well documented examples. So I’m writing about it here.. <a href="https://towardsdatascience.com/drawing-from-a-random-distribution-in-sql-b6c97c89e45c">Read More</a>