Contact Information:

email: arostami (at) eecs (dot) berkeley (dot) edu

office: UC Berkeley

Sutardja Dai Hall

Berkeley, CA 94720

Afshin Rostamizadeh



*NIPS 2011 Workshop*: Domain Adaptation

I am now at Google Research NY!


Intro:

I am currently a postdoc in Peter Bartlett's group at the Computer Science department of UC Berkeley.

My CV.

Education:

I completed my PhD studies in the Computer Science department of the Courant Institute of Mathematical Science at New York University under the generous guidance of Mehryar Mohri.

I've actually been at UC Berkeley before, recieving my Bachelor of Science in Electrical Engineering and Computer Science.

Research Interest:

I am interested in various Machine Learning Theory problems (with an eye on applications). I've worked on problems regarding learning from non-I.I.D. distributions, learning from biased distributions, adaptation and automatically learning with multiple kernels.

Teaching:

I have TA'ed Data Structures and Algorithms and Foundations of Machine Learning.

Publications:

- Learning with Missing Features, Afshin Rostamizadeh, Alekh Agarwal and Peter Bartlett, UAI 2011

- Ensembles of Kernel Predictors, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, UAI 2011

- Matrix Coherence and the Nystrom Method, Ameet Talwalkar and Afshin Rostamizadeh, UAI 2010

- Generalization Bounds for Learning Kernels, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, ICML 2010

- Two-Stage Learning Kernel Algorithms, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, ICML 2010

- Stability Bounds for Stationary phi-mixing and beta-mixing Processes, Mehryar Mohri and Afshin Rostamizadeh, JMLR 2010, Volume 11

- Learning Non-Linear Combinations of Kernels, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, NIPS 2009

- Domain Adaptation: Learning Bounds and Algorithms, Yishay Mansour, Mehryar Mohri and Afshin Rostamizadeh, COLT 2009

- L2 Regularization for Learning Kernels, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, UAI 2009

- Multiple Source Adaptation and the Renyi Divergence, Yishay Mansour, Mehryar Mohri and Afshin Rostamizadeh, UAI 2009

- Rademacher Complexity Bounds for Non-i.i.d. Processes, Mehryar Mohri and Afshin Rostamizadeh, NIPS 2008

- Learning Sequence Kernels, Corinna Cortes, Mehryar Mohri and Afshin Rostamizadeh, MLSP 2008 (invited lecture)

- Domain Adaptation with Multiple Sources, Yishay Mansour, Mehryar Mohri and Afshin Rostamizadeh, NIPS 2008

- Sample Selection Bias Correction Theory, Corinna Cortes, Mehryar Mohri, Michael Riley and Afshin Rostamizadeh, ALT 2008

- Stability Bounds for Non-i.i.d. Processes, Mehryar Mohri and Afshin Rostamizadeh, NIPS 2007 [poster]

- Balancing Traffic Load in Wireless Networks with Curveball Routing, Lucian Popa, Afshin Rostamizadeh, Richard M. Karp, Christos Papadimitriou and Ion Stoica, MobiHoc 2007

Other Work:

- Kernel Learning: Automatic Selection of Optimal Kernels, NIPS 2008 Workshop, co-organized with Corinna Cortes, Arthur Gretton, Gert Lanckriet and Mehryar Mohri

- Learning Kernels, ICML 2011 Tutorial, with Corinna Cortes and Mehryar Mohri