## Overview

I am a PhD student at the University of Washington, working with Carlos Guestrin on problems in machine learning.

In 2012, I received bachelor's degrees in electrical engineering and computer science from the University of Michigan in Ann Arbor. At Michigan, I also worked on some research projects with Clayton Scott.

## Recent News

- Next summer I will be working as an autonomous vehicles intern at General Motors in my home state of Michigan.
- In December, I attended NIPS in Montreal.
- In October, I presented my work on Blitz at Facebook.

## Research

I research methods for large-scale machine learning. I am interested in how existing algorithms can be modified to reduce computation. Currently I am looking at several ideas for this in the context of sparse optimization.

## Publications

Unified Methods for Exploiting Piecewise Structure in Convex Optimization.

Advances in Neural Information Processing Systems 29, 2016.

(To appear)

Blitz: A Principled Meta-Algorithm for Scaling Sparse Optimization.

International Conference on Machine Learning, 2015.

author = {Tyler B. Johnson and Carlos Guestrin},

title = {Blitz: A Principled Meta-Algorithm for Scaling Sparse Optimization},

booktitle = {International Conference on Machine Learning},

year = {2015}

}

## Software

A Python package implementing the Blitz algorithm for sparse regression can be found on GitHub here.

Currently I do not have plans to release limited-memory and distributed implementations as libraries. For those, you can see the code for my experiments in the Blitz paper, which will be posted soon.

## Teaching

I have enjoyed helping teach the following courses:

- EE 578: Convex Optimization, UW, Winter 2014
- CSE 546: Machine Learning, UW, Fall 2013
- ENGR 100: Music Signal Processing, UM, Fall 2010
- EECS 314: Electronic Circuits, UM, Winter 2010