— In the News —
Here's how a math trick that's commonly used in sports science to find "meaningful results" in small sample sizes is seriously flawed. Even so, it's widely used and the leading paper promoting the technique has more than 2500 citations. This article from FiveThirtyEight explores the method and how it's managed to thrive in spite of its problems.
More and more, cross-border data flows are key to the global economy and yet, governments are increasingly blocking those flows. Some of it's about privacy but it's far from just that. This article in The Financial Times takes a hard look at what's at stake and what issues need to be navigated.
— Profiles —
Judea Pearl, a pioneering figure in artificial intelligence, argues that AI has been stuck in a decades-long rut. His prescription for progress? Teach machines to understand the question why.
— Sponsored Link —
The AI Conference is where cutting-edge science meets new business implementation. It's a deep dive into emerging AI techniques and technologies with a focus on how to use it in real-world applications. Presented by O'Reilly Media and Intel. Save 20% on most passes when you use the discount code ELIXIR20. Last year this event sold out. Don't miss out, register today.
— Tools and Techniques —
Machine learning isn't prohibited but the GDPR impacts machine learning in key ways. Here's a thoughtful set of answers to commonly asked questions about maintaining machine learning programs that are GDPR-compliant.
Great interactive article from the Berkeley AI Research group that shows how "fair" machine learning algorithms can have unintended consequences over the long term.
This is the beginning of what looks like will be a fantastic series about building a data science discipline at a startup. The latest installment covers considerations for building data pipelines, including the evolution of data pipelines, properties of a high functioning data pipeline, and a sample pipeline built on GCP. For information about the entire series, see the Introduction.
This is a great tutorial that shows how to start and finish a simple data science project using Python and Yellowbrick - from exploratory data analysis, to feature selection and feature engineering, to model building and evaluation.
This set of TensorFlow tutorials combines detailed tutorials with simple code implementations.