Welcome to my home page!
|
Hi. My name is Damian Eads. I am a third year Ph.D. student in Computer Science at the University of California, Santa Cruz. I may be e-mailed at eads at soe dot ucsc dot edu. |
As a Graduate Student Researcher (GSR) at UCSC, I conduct research in machine learning and time series classification under the direction of Dr. David Helmbold at UC Santa Cruz and Dr. James Theiler at Los Alamos National Laboratory. In addition, I also collaborate with Dr. Ethan Miller at UC Santa Cruz on the application of machine learning algorithms to storage and operating systems problems and Dr. Joshua Bloom at UC Berkeley on its application to astrophysics.
As a Staff Research Assistant at Los Alamos National Laboratory, I work for the Intelligent Searching of Images and Signals team, part of the International, Space and Response division, formerly Nonproliferation and International Security Division. When at Los Alamos, I conduct research in machine learning applied to geospatial, remote sensing, and time series problems. Los Alamos is kind of like a magnet; I've worked there since 2001, and I keep on coming back.
From mid-June of 2008 to late September 2008, I will be conducting my research on the LANL campus. My e-mail there is eads@lanl.gov, and my office phone number, 505-667-3659.
In May of 2004, I earned my Bachelor's Degree (w/ High Honors) in Computer Science at the Rochester Institute of Technology in Rochester, NY, minoring in Mathematics and Psychology.
Jordan Johnson and I studied multiple dispatch in our Programming Languages and Semantics class. We had a lot of fun, learned quite a bit, and wrote an interesting report.
I took Combinatorics with Professor Paul Wilson at RIT. It was by far, the most mind-expanding course I have ever taken. Jon Coles and I used to abuse our brains cranking on problems for many hours each week at MacGregor's. After finishing his Master's Thesis on Ramsey and Folkmann numbers, Jon moved to Zurich to pursue a doctorate in Physics. Go Jon!
In my spare time, I manage a new open source software project called the Chestnut Scientific Library. Most of the core algorithms are implemented in C and are interfaced with Python. The new package will require Numpy. Some of the algorithms implemented include:
- boosting: AdaBoost, LPBoost, BrownBoost, MADABoost, SoftBoost, TotalBoost
- hidden markov models with support for discrete and continuous emission distributions and multiple training sequences.
- tools for basic time series analysis
- clustering: hierarchical, k-means, QT, and shifting means algorithms
- spatial tools: Voronoi tesselations, kd-trees, k-nearest neighbor
- basic discriminant analysis: FLD, LDA, QDA
The primary purpose of this site is professional so there is not much that's personal here. The best way to get to know me is in person.
Check out a small slice of my interesting ancestral history.
Thanks to Andreas Viklund for the wonderful web page design.