Some Technical Reading

I am ever-optimistic that on the weekend I will have enough time and brainpower to do some serious technical reading. The following are five articles on my list. I hope that at least one of them resonates with you too. Happy reading.

1. In this paper from Nature Communications, Kun-Hsing Yu et al. use machine learning to diagnose cancer. The authors write:

We extract 9,879 quantitative image features and use regularized machine-learning methods to select the top features and to distinguish shorter-term survivors from longer-term survivors with stage I adenocarcinoma (P<0.003) or squamous cell carcinoma (P=0.023) in the TCGA data set.

Figure a shows ROC curves for classifying lung adenocarcinoma versus adjacent normal tissue.

1. John Mount and Nina Zumel describe Why you should re-encode high dimensional categorical variables.

2. Jamie Ashander delves into the details of the statistics of General Linear Mixed Models, and provides the R code to help you build your own models.

3. Sebastian Ruder does a fine job of explaining the math in his post An overview of gradient descent optimization algorithms.

4. Iwana and Uchida use deep learning to do something your mother told you never to do: Judging a Book by Its Cover.