The BG/NBD model explained. — Knowing how much value each particular customer adds to your business is crucial, not least because it helps define the customer acquisition and retention budgets. Customer Lifetime Value models are a useful tool for gaining such insights. There seems to be, however, a lot of confusion about them, and apparently…

How to make probabilistic comparisons that embrace uncertainty — Comparing things is hard. Even more so, when the information the comparison is based on is only partial or uncertain. Beware those who offer single numbers which, they claim, harness and capture all the uncertainty. Luckily, Bayes’ got you covered. …

Always know what to expect from your data — This article was first published on Neptune AI’s blog. You are a part of a data science team at a product company. Your team has a number of machine learning models in place. Their outputs guide critical business decisions, as well as a couple of dashboards displaying important KPIs that…

Because GANs are not all you need — When you hear about computer-generated images, you probably think about deep fakes, the cats that don’t exist, or a horse turned zebra. And this, quite reasonably, brings up GANs to your mind. But even though these models have seen great success in generating realistic images, they come with their shortcomings…

Learn a couple of useful probabilistic tricks to safely navigate the sea of incomplete information — Sometimes we wish we knew something that we don’t. Unfortunately, in many cases, there is no time or even no way to learn what we need. Nevertheless, decisions and assessments need to be made with only disposable knowledge. While to many of us navigating in the mist of incomplete information…

Revolutionizing a simple GAN to improve the quality of generated images. — In my previous article on Generative Adversarial Networks, we have seen how these ingeniously simple yet powerful models work, what they can be used for, and we have built a very simple GAN for generating new species of Pokemons. Check it out if you haven’t read it yet: Generative Adversarial Networks 101 How to build a simple GANtowardsdatascience.com

Keeping accuracy high while getting suggestive explanations that create knowledge and help understand and debug data. — Microsoft Research has recently developed a new boosting-based model which they claim yields as accurate predictions as state-of-the-art methods while providing an innovative way to understand its workings. The Explainable Boosting Machine, for that’s what it’s called, is unique in how it delivers new knowledge and allows to debug and…

This simple mind-shift will help you better understand the uncertain world around you — Human brains don’t process probabilities very well. We play lotteries but are afraid to board a plane. There are biological underpinnings to this: overestimating small probabilities that would be disastrous should they materialize has helped our ancestors survive. In today’s world, however, being able to think in a statistically sound…