I retired from my role as a Partner Researcher and Research Manager at Microsoft Research in October of 2021 after 25+ years. I am currently an affiliate professor in the Department of Statistics at the University of Washington. At Microsoft, I worked with many people across Microsoft on a wide range of projects. During my tenure at Microsoft, machine learning changed from an esoteric area of research to a core software technology. I am incredibly fortunate to have been involved in this change, bringing machine learning algorithms to many products and services at Microsoft including Windows, SQL Server, and Bing to name just a few. Some of the technical innovations that I have been involved in are captured in published research papers. Here is a list of papers available at my former employer. If you are interested in a more complete list of papers you can use the Google Scholar Profile link below. If you would like to contact me, please use my personal email address or contact me via LinkedIn. Below are a few things that amuse me.

Impossible Things

Puzzles

A geometric puzzle

Find the counterfeit coin

You are given 12 gold coins that appear to be identical in size. One of the coins is a counterfeit and is either heavier or lighter than the rest. You need to identify the counterfeit coin using a balance scale. Can you find the counterfeit using only three measurements? Can you do 9 coins in two measurements?

Random Streaming Selection Algorithms

You are given access to an enumeration of items and asked to choose one item at random. You have access to a random number generator but do not know the number of items in the enumeration, can only sequentially access items, and can only store a single item in memory. How can you choose a random item? Can you do it with far fewer calls to the random number generator than there are items in the enumeration? What if you want to choose k items at random from an enumeration?

How many faces can you find?

Illusions

Images

Photos