I am a Ph.D. student in the Robotics Institute at the School of Computer Science in Carnegie Mellon University.  My advisor  is Geoffrey J.  Gordon.  I am a member of the SELECT Lab, which is headed by Geoffrey Gordon and Carlos Guestrin. I used to work with Andrew Moore, and I am also affiliated with the AUTON Lab.

I am interested in efficient inference, learning and model selection for low-dimensional representations of dynamical systems, including latent variable models and predictive state representations, from high-dimensional data.  The algorithms I work on require tools from data mining, statistical machine learning, convex optimization, matrix analysis and control theory.  I apply my research to sequential data modeling and prediction problems in robot cognition, computer vision and biosurveillance.

Before graduate school, I studied Computer Science, Mathematics and Economics at the University of Southern California in Los Angeles, where I worked with Gaurav Sukhatme at the Robotics and Embedded Systems Lab. Prior to that, I went to school in Jeddah, Saudi Arabia. I'm originally from  India, however, and my native language is Urdu.

Here's my academic CV, my research statement, and resume.

Email: 
siddiqi at cs dot cmu dot edu

Office:
Newell-Simon Hall (NSH) 3122

Mailing Address:
5000 Forbes Ave.
NSH 3122, Robotics Institute
Carnegie Mellon University
Pittsburgh, PA 15213

Updates:
10/15/09: Defended thesis!
10/6/09:  Tech report on 'Reduced-Rank HMMs'  now on arXiv
8/20/09: Defending my thesis October 15th, looking for jobs!
12/9/08:  Initial version of C code for HMM learning with STACS and EM now                            online (Linux binaries and scripts)