Computer Science Engineer

Saurabh Paul

Amos Eaton ***, o ce: *** 518-***-****

Information Dept of Computer Science, mobile: 001 518-***-****

Rensselaer Polytechnic Institute, e-mail: ******@**.***.***

*** *** ******, ***y, NY 12180-3590,USA. http://www.cs.rpi.edu/~pauls2

To secure a summer internship in a research lab or software engineering industry.

Objective

Machine Learning : Dimensionality reduction, Large-Scale Learning, Low-rank approximations, Clas-

Research

Interests si cation, Clustering.

Others : Randomized Algorithms for numerical linear algebra.

Rensselaer Polytechnic Institute, Troy, NY, USA

Education

Ph.D. in Computer Science August 2010 May 2015

GPA : 3.87/4.0

Adviser: Professor Petros Drineas

Rensselaer Polytechnic Institute, Troy, NY, USA

M.S. in Computer Science August 2010 Dec 2012

GPA : 3.98/4.0

Adviser: Professor Petros Drineas

Bengal Engineering and Science University, Shibpur, India

B.E. (Hons) in Computer Science and Technology July 2006 May 2010

Saurabh Paul, Christos Boutsidis, Malik Magdon-Ismail and Petros Drineas.

Manuscripts

Random Projections for Support Vector Machines.

Saurabh Paul, Ke Huang, Nathan Kupp, Petros Drineas and Yiorgos Makris.

Dimensionality Reduction for Accelerated Evaluation and Compaction of Machine Learning-Based

Analog/RF Tests.

Rensselaer Polytechnic Institute, Troy, NY, USA

Academic

Experience

Research Assistant Jan 2011 present

Rensselaer Polytechnic Institute, Troy, NY, USA

Teaching Assistant for Operating Systems. Aug 2010 Dec 2010

Fast Algorithm for Quadratic Programming Feature Selection

Research

Projects Adviser: Dr. Petros Drineas

Investigating ways to speed up quadratic programming feature selection for large-scale machine learn-

ing problems. Support Vector Machines will be used for comparing classi cation accuracy.

A Linear Algebraic approach to Electronic Circuit Testing

Adviser: Dr. Petros Drineas

Investigating linear algebraic methods of dimensionality reduction like Singular Value Decomposition

and Random Projections on large scale circuit-testing datasets provided by IBM and Texas Instru-

ments. Used Support Vector Machines and Nearest Neighbors to improve test-escapes and yield-loss.

Random Projections for Support Vector Machines

Adviser: Dr. Petros Drineas

The linear support vector machine constructs a hyperplane separator that maximizes the 1-norm soft

margin. We develop a new oblivious dimension reduction technique which is precomputed and can

be applied to any input data matrix. We prove that, with high probability, the margin and minimum

enclosing ball in the feature space are preserved to within small relative error, ensuring comparable

generalization as in the original space. Extensive experiments on real-world data support our theory.

Bagging for Improved Performance August 2009 May 2010

Undergraduate

Projects Bengal Engineering & Science University, Shibpur, India

Used bagging and an ensemble of classi ers to obtain improved classi cation accuracy on various

small and medium-sized datasets. Codes were written in C.

Design of File Transfer Application June 2009 July 2009

Indian Statistical Institute, Kolkata, India.

A File Transfer application with encryption facility using RSA and user interface using Gnome Toolkit

was built.

Automatic Analysis of PET Tumor Images for Radiotherapy Treatment June 2008

Queen s University, Belfast, UK.

Developed an automatic image classi cation system that can classify tumors on the basis of shape

and size and also output tumor position in the image. Codes were written in Matlab.

Computer Operating Systems, Programming Languages, Computer Algorithms, Computability and

Graduate

Coursework Complexity, Machine Learning, Database Mining, Linear Algebra, Randomized Algorithms, Compu-

tational Optimization, Computational Linear Algebra.

Programming Languages : C, C++, Python, MATLAB, HTML.

Computer Skills

Database Management Systems: SQL.

Tools: LIBSVM, CLapack, Gnome Toolkit.

Operating Systems: Linux/Unix, Windows.

Typography: Latex, Microsoft O ce.

Received Full Scholarship for Electronics Engineer s Welcome Scheme from Queen s University,

Academic

Honors Belfast, 2008.

Placed within 0.57% among 60,000 candidates in WBJEE (Engineering Entrance Exam) 2006.

Secured 25th Rank among over 400,000 students in Madhyamik Pariksha (Secondary School

Exam), 2004.

Introducing Prospective Graduate Students to di erent research groups at CS Dept, RPI.

Appointments &

Department Peer advisor to incoming graduate students for Fall 2011, 2012.

Service Member of Graduate Admissions Committee, RPI, USA since Fall 2011.

ACM Student Member.

Contact this candidate