Functional Feature Construction for Individualized Treatment Regimes

Consistent Individualized Feature Attribution for Tree Ensembles

Interpreting predictions from tree ensemble methods such as gradient boosting machines and random forests is important, yet feature attribution for trees is often heuristic and not individualized for each prediction. Here we show that popular feature attribution methods are inconsistent, meaning they can lower a feature’s assigned importance when the true impact of that feature actually increas...

Individualized Treatment for Allergic Rhinitis

There Is Individualized Treatment . Why Not Individualized

Doctors use statistics to advance medical knowledge; we use a medical analogy to introduce statistical inference “from scratch” and to highlight an improvement. Your doctor, perhaps implicitly, predicts the effectiveness of a treatment for you based on its performance in a clinical trial; the trial patients serve as controls for you. The same logic underpins statistical inference: to identify t...

Feature Construction for Game Playing1

To build an evaluation function for game-playing, one needs to construct informative features that enable accurate relative assessment of a game state. This chapter describes the feature construction problem, and suggests directions for dealing with shortcomings in the present state of the art.

Genetics and Individualized Treatment for Colorectal Cancer

The clinical benefit of the genomics revolution is particularly evident in oncology. Genetic characterization of patients’ tumors has revealed that cancer is fundamentally a heterogeneous disease whose complex clinical course can be correlated to specific molecular profiles. Now, genomic data and tools are beginning to be translated from the laboratory to the clinic where physicians and patient...