Monthly Archives: December 2020

Why COVID-19 is Overrunning the US in Late 2020: Overlapping Epicurves

Data in simulated epicurves show frequencies and explain outbreak timing

While other countries have found a way to control their community spread of COVID-19 while waiting for the vaccine program to be implemented, the United States has totally failed at this. An epicurve is a diagram of the timing of an outbreak, and in other countries, this curve has been flattened. But in the United […]

This Course in Explainable AI will Get you Ready for the Future!

What do the data say when a machine learning algorithm is applied, and which features are important?

We experience artificial intelligence all the time on the internet in terms of friend suggestions on social media, internet ads that reflect what we have been searching for, and “smart” recommendations from online stores. But the reality is that even the people who build those formulas cannot usually explain why you were shown a certain […]

Read Our New Peer-reviewed Paper on the Ketogenic Hypothesis for Lipedema!

Lipedema is a chronic condition that is often misdiagnosed as obesity

Lipedema, a severe metabolic disorder, is more common than originally thought. A non-trivial proportion of women who struggle with obesity actually have undiagnosed lipedema. I am on a research team that just published a peer-reviewed article that presents the ketogenic hypothesis for lipedema, and here, I present a summary.

Two Takeaways from Danny Ma’s Machine Learning Panel: Understanding the Problem, and Understanding your Data

Roller coaster like an ETL pipeline that does automation

This lively panel discussed many topics around designing and implementing machine learning pipelines. Two main issues were identified. The first is that you really have to take some time to do exploratory research and define the problem. The second is that you need to also understand the business rules and context behind the data.