MANU NANDAN
*******@***.*** *** University Village apt. 2
http://www.cise.ufl.edu/~mnandan/ Gainesville, FL-32603
http://www.linkedin.com/pub/manu-nandan/11/645/90b Phone: 352-***-****
OBJECTIVE
Seeking a full time job in the machine learning industry, where I can apply and enrich my
research expertise.
EDUCATION
PhD University of Florida, Computer Engineering (GPA: 3.69) Dec 2013
Advisor: Dr. Pramod Khargonekar
GPA on Machine learning courses (7): 3.88
MS University of Florida, Computer Engineering (GPA: 3.68) Dec 2009
BTech National Institute of Technology Calicut, India May 2003
Electronics and Communications Engineering (69%)
RESEARCH AND PUBLICATIONS
• Developed a new SVM objective for fast training and classification of large datasets [2,3]
o Proposed approximate extreme points SVM (AESVM) that approximately solves
SVM for very large datasets
o Proposed a linear time algorithm to pre-process the data for AESVM
o Derived theoretical and empirical results that guarantee that AESVM and SVM
solutions are similar
o Extended the algorithm to achieve sparse SVM outputs leading to fast
classification
• Developed an SVM variant for datasets that are too large to fit in memory [4]
o Heuristic algorithm using convex optimization techniques
o Gives approximate solutions with one read of data from HDD and accurate
solutions with two reads
• Currently developing a parallel SVM variant for text classification
o Fast algorithm using stochastic gradient descent and coordinate descent [4]
o Very low communication overhead makes it ideal for implementation on
MapReduce
Manu Nandan - 1
• Researched application of variants of support vector machines (SVMs) to seizure onset
detection [1,A]
o Implemented and compared algorithms for anomaly detection
o Performed time series analysis using wavelet decomposition of EEG
• Researched application of hidden Markov models (HMMs) to seizure detection
Journal and Conference papers:
[1] Manu Nandan, Sachin S. Talathi, Stephen Myers, William L. Ditto, Pramod P.
Khargonekar, and Paul R. Carney, Support vector machines for seizure detection in an
animal model of chronic epilepsy, Journal of Neural Engineering, June 2010
[2] Manu Nandan, Pramod P. Khargonekar, and Sachin S. Talathi, Fast SVM Training Using
Approximate Extreme Points, Journal of Machine Learning Research, under review
(accepted with minor revisions), draft version: http://arxiv.org/abs/1304.1391
[3] Manu Nandan, Pramod P. Khargonekar, and Sachin S. Talathi, Real time epileptic seizure
onset detection using approximate extreme points SVM, Appearing in the ICML 2013
Workshop on Role of Machine Learning in Transforming Healthcare
[4] Manu Nandan, Pramod P. Khargonekar, and Sachin S. Talathi, Text classification using
approximate extreme points SVM, under preparation
Posters:
[A] Sachin S. Talathi, Manu Nandan, William L. Ditto, Pramod P. Khargonekar, and Paul R.
Carney, Support vector machine algorithms for early seizure detection in an animal model of
temporal lobe epilepsy, Annual American Epilepsy Society Meeting, Boston, Dec (2009)
[B] Manu Nandan, Pramod P. Khargonekar, and Sachin S. Talathi, Fast and robust offline
epileptic seizure detection using Rossler oscillators, The 6th International Workshop on
Seizure Prediction, Nov (2013)
EMPLOYMENT HISTORY
Research Assistant at the University of Florida 2008-Present
• Research on machine learning and its application to large datasets
• Application of machine learning to seizure detection and prediction
• High performance implementation of convex optimization methods
Advanced software engineer, Delphi 2003-2007
• Development and testing of automotive embedded software in C and C++
• Design and development of automotive signal simulator hardware using
microprocessors
• Development of Perl scripts for task automation
Manu Nandan - 2
AWARDS
Quality Performance Award, 2005, at Delphi (included a cash price of INR 10000)
Nominated by the University of Florida for the Howard Hughes Medical Institute
international student research fellowship in 2012
PROFESSIONAL SERVICE
Peer-Reviewed Articles for Computers in Biology and Medicine, 2010-Present
COMPUTER SKILLS
Programming: C, C++, MATLAB, Python, Perl, Java, MySQL, MPI
Platforms: Linux, Windows
REFERENCES
Dr. Pramod Khargonekar (***@***.***), Professor and Eckis Chair, Dean Emeritus,
Department of Electrical and Computer Engineering, University of Florida, Gainesville
(also Assistant director for the Directorate of Engineering at National Science Foundation)
Dr. Sachin Talathi (*******@***.***), Assistant Professor, Dept of Biomedical Engineering,
University of Florida, Gainesville
Dr. Paul Gader (******@****.***.***), Professor and Interim Chair, Dept of Computer and
Information Science and Engineering, University of Florida, Gainesville
Manu Nandan - 3