Garvesh Raskutti's Webpage
Garvesh Raskutti
I am a PhD candidate in statistics
at UC Berkeley,
where I started in the fall of 2007. I work under the joint supervision of Bin Yu and
Martin
Wainwright. Before this, I was an undergrad and a masters
in electrical engineering student at University of Melbourne, Australia working with Rodney Tucker and
Kerry Hinton.
Email: garveshr@stat.berkeley.edu
CV:pdf
Publications
Theoretical Statistics and Machine Learning
Early stopping of gradient descent over kernel classes: An optimal data-dependent stopping rule.,
Garvesh Raskutti, Martin Wainwright and Bin Yu Allerton 2011.
Minimax-optimal rates for sparse additive models over kernel classes via convex programming,
Garvesh Raskutti, Martin Wainwright,
and Bin Yu. Submitted to Journal of Machine Learning Research . Preliminary version in NIPS, 2009.
[pdf]
Efficient
Restricted Eigenvalue Properties for Correlated Gaussian Designs,
Garvesh Raskutti, Martin Wainwright, and Bin Yu. Journal of Machine Learning Research 2010.
Minimax rates of estimation for high-dimensional linear regression over `l_q-balls,
Garvesh Raskutti, Martin Wainwright,
and Bin Yu. IEEE Transactions on Information Theory, To appear.
High-dimensional covariance estimation by minimizing `l_1-penalized log-determinant
divergence,
Pradeep Ravikumar, Martin Wainwright,
Garvesh Raskutti, and Bin Yu. Electronic Journal of Statistics .
Comments on Envelope Models,
Jinzhu Jia, Yuval Benjamini, Chingway Lim, Garvesh Raskutti and Bin Yu . Statistica Sinica 2010 .
Model Selection in Gaussian Graphical Models: High-Dimensional Consistency of l_1-
regularized MLE,
Pradeep Ravikumar, Garvesh Raskutti, Martin Wainwright and Bin Yu NIPS 2008.
Optical Network Design
A New Method for Blocking Probability Evaluation in OBS/OPS Networks with Deflection Routing ,
Eric M. W. Wong, Jayant Baliga, Moshe Zukerman, Andrew Zalesky,
and Garvesh Raskutti. IEEE Journal of Lightgwave Technology 2009.
Switching energy and device size limits on digital photonic signal processing technologies,
Kerry Hinton, Garvesh Raskutti,Peter Farrell, and Rodney Tucker.
IEEE Journal of Selected Topics in Quantum Electronics. Preliminary version at COIN-ACOFT 2007.
Blocking Probability Estimation for Trunk Reservation Networks,
Garvesh Raskutti, Andrew Zalesky, Eric Wong and Moshe Zukerman. International Conference on Communications
2007 . Preliminary version in IEEE Communications Letters 2007 .
[pdf]
Energy consumption limits in high-speed optical and electronic signal processing,
Rodney S. Tucker, Kerry Hinton and Garvesh Raskutti. Electronics Letters August 2007 .
Statistical Genomics
Pretreatment gene expression profiles can be used to predict response to neoadjuvant
chemoradiotherapy in esophageal cancer,
Cuong Duong, Danielle Greenawalt, Adam Kowalczyk, Marianne Ciavarella, Garvesh Raskutti,
William Murray, Robert Thomas and Wayne A. Phillips.
Annals of Surgical Oncology, December 2007 .
Large Validation of Anti-learnable Signature in Classi cation of Response to Chemora-
diotherapy in Esophageal Adenocarcinoma Patients,
Adam Kowalczyk, Danielle M. Greenawalt,
Justin Bedo, Cuong Duong, Garvesh Raskutti, Robert J. S. Thomas, Wayne A. Phillips.
Conference on Optimization and Systems Biology, August 2007 .
Presentation Slides
Minimax rates for sparse additive models over kernel classes, presented at JSM 2011 (Miami) and WITMSE 2011 (Helsinki)
[pdf]
Slides on Markov Logic Networks, presented at Microsoft Research Asia, 2010
[pdf]
High-dimensional regression under $\ell_q$-ball sparsity: Optimal rates of convergence , presented at Allerton, 2009.
[pdf]
TA Experience
Stat 20, Introductory Statistics, UC Berkeley, Fall 2011
Stat 210B, Graduate Theoretical Statistics, UC Berkeley, Spring 2011
Stat 131A, Statistics for Social Sciences, UC Berkeley, Fall 2009
EE 325, Stochastic Signals and Systems, University of Melbourne, 1st Semester 2007
EE 221, Fundamentals of Signals and Systems, University of Melbourne, 2nd Semester 2006
Classes Taken
Probability, Theoretical Statistics, Applied Statistics, Graphical Models, Statistical Learning Theory, Convex Optimization, Information Theory and Statistics, Bayesian Statistics, High-Dimensional Statistics, Statistical Consulting.