BIU learning club – Shalev Shaer – Betting as a mechanism to make reliable discoveries

Zoom link: 

https://biu-ac-il.zoom.us/j/4685913265

 Title:

Betting as a mechanism to make reliable discoveries 

Abstract:

This talk introduces a new statistical testing framework that allows researchers to analyze an incoming i.i.d. data stream with any arbitrary dependency structure, and safely conclude whether a feature is conditionally associated with the response under study. We allow the processing of data points online, as soon as they arrive, and stop data acquisition once significant results are detected, rigorously controlling the type-I error rate. Our test can work with any sophisticated machine learning algorithm to enhance data efficiency to the extent possible. We will demonstrate the practicality of our proposal by applying it to carefully-designed simulated data, and to a real-world task of detecting mutations that developed resistance to an HIV drug.

Short Bio:

Shalev Shaer is a 3rd year PhD student at the Technion in the ECE faculty under the supervision of Prof. Yaniv Romano. His primary research interest lies at the intersection of statistics and ML, with a focus on online testing using test martingales and the integration of these concepts within ML and vice versa. Prior to that, he graduated (summa cum laude) from the ECE faculty at the Technion. In addition to his academic pursuits, he is also a research intern at Microsoft, where he has the opportunity to work on developing speech and computer vision algorithms.