Experience
Sign in to view more
Education
Sign in to view more
Bio
I am currently working at the Department of Computer Science and Engineering, IIT Kanpur as an assistant professor. I am a Dr. Deep Singh and Daljeet Kaur Faculty Fellow at the department.
Research Interests
I am broadly interested in machine learning with special emphasis on online and stochastic learning and optimization, extreme multi-label classification, high dimensional statistics, and kernel methods. I also maintain an avid interest in statistical learning theory, complexity theory, data streaming algorithms, and computational geometry.
Employment
I was previously working at Microsoft Research India as a Post-doctoral Researcher working with the Machine Learning, Natural Language Systems and Applications (MLN) Group.
Employment History
Assistant Professor, Department of CSE, IIT Kanpur, September 2015 - present.
Post-doctoral Researcher, Microsoft Research India Pvt. Ltd., Jul 2013 - August 2015.
Doctoral Research Intern, Microsoft Research India Pvt. Ltd., May 2012 - Aug 2012.
Education
I was a Ph.D. student at the Department of Computer Science and Engineering, IIT Kanpur working under the joint supervision of Prof. Harish C. Karnick and Prof. Manindra Agrawal. During this period I also worked closely with Dr. Prateek Jain (Microsoft Research India) and Prof. Bharath Sriperumbudur (Pennsylvania State University).
Educational Qualifications
Doctor of Philosophy in Computer Science and Engineering, IIT Kanpur, 2008-2013 (degree awarded 2014), CPI - 9.60.
Bachelor of Technology in Computer Science and Engineering, IIT Kanpur, 2004-2008, CPI - 9.90.
All India Senior School Certificate Examination, Bal Bharati Public School, 2004, Aggregate - 93.4%.
All India Secondary School Examination, Bal Bharati Public School, 2002, Aggregate - 94.8%.
Selected Publications (complete list available on my webpage)
◦ Robust Regression via Hard Thresholding. Kush Bhatia, Prateek Jain, and Purushottam Kar, 29th Annual
Conf. on Neural Information Processing Systems (NIPS), 2015.
◦ Sparse Local Embeddings for Extreme Multi-label Classification. Kush Bhatia, Himanshu Jain, Purushottam
Kar, Prateek Jain, and Manik Varma, 29th Annual Conf. on Neural Information Processing Systems (NIPS),
2015.
◦ Surrogate Functions for Maximizing Precision at the Top. Purushottam Kar, Harikrishna Narasimhan, and
Prateek Jain, 32nd International Conference on Machine Learning (ICML), 2015.
◦ Optimizing Non-decomposable Performance Measures: A Tale of Two Classes. Harikrishna Narasimhan,
Purushottam Kar, and Prateek Jain, 32nd International Conference on Machine Learning (ICML), 2015.
◦ Online and Stochastic Gradient Methods for Non-decomposable Loss Functions. Purushottam Kar, Harikrishna
Narasimhan, and Prateek Jain, 28th Annual Conf. on Neural Information Processing Systems (NIPS), 2014.
◦ On Iterative Hard Thresholding Methods for High-dimensional M-Estimation. Prateek Jain, Ambuj Tewari,
and Purushottam Kar, 28th Annual Conference on Neural Information Processing Systems (NIPS), 2014.
◦ Large-scale Multi-label Learning with Missing Labels. Hsiang-Fu Yu, Prateek Jain, Purushottam Kar, and
Inderjit S. Dhillon, 31st International Conference on Machine Learning (ICML), 2014.
◦ On the Generalization Ability of Online Learning Algorithms for Pairwise Loss Functions. Purushottam Kar,
Bharath Sriperumbudur, Prateek Jain, and Harish Karnick, 30th International Conference on Machine Learning
(ICML), 2013.
◦ Supervised Learning with Similarity Functions. Purushottam Kar and Prateek Jain, 26th Annual Conference
on Neural Information Processing Systems (NIPS), 2012.