Research Interests
My research interest primarily is in the area of Machine Learning, with focus on Kernel methods. The main theme in most of my works is formalizing novel learning settings as optimization problems that have theoretical guarantees and can be applied to diverse applications. The key tools that I leverage in the context of learning are Optimization and Statistics. I am currently pursuing projects on i) Optimal Transport ii) Fujitsu sponsored project on Causality iii) other collaborations with folks in Microsoft, IITB, IITM, IISc. Link to my lab: ML LAB.
Publications
Pre-prints
Conference Proceedings
- Piyushi Manupriya, Pratik Kumar Jawanpuria, Karthik Gurumoorthy, Sakethanath Jagarlapudi and Bamdev Mishra. Submodular Framework for Structured-sparse Optimal Transport . Accepted at ICML-2024. pdf
- Piyushi Manupriya, Rachit Keerti Das, Sayantan Biswas, and J. Saketha Nath. CONSISTENT OPTIMAL TRANSPORT WITH EMPIRICAL CONDITIONAL MEASURES . Accepted at AISTATS-2024. pdf
- Piyushi Manupriya, SakethaNath Jagarlapudi, Tarun Ram Menta and Vineeth N. Balasubramanian. Improving Attribution Methods by Learning Submodular Functions. Accepted at AISTATS'22. |paper||poster||slides||Talk||Code|
- Soumya Chatterjee, Ayush Maheshwari, Ganesh Ramakrishnan and SakethaNath Jagaralpudi. Joint Learning of Hyperbolic Label Embeddings for Hierarchical Multi-label Classification. Accepted at EACL'21, pdf
- J. Saketha Nath and Pratik Jawanpuria. Statistical Optimal Transport posed as Learning Kernel Embedding. Accepted at NeurIPS'20. Long pdf, NeurIPS pdf, 3min Video,
- Prafull Prakash, Chaitanya Murti, J. Saketha Nath and Chiranjib Bhattacharyya. Optimizing DNN Architectures for High Speed Autonomous Navigation in GPS Denied Environments on Edge Devices. PRICAI (2) 2019.
- Ayush Maheshwari, Vishwajeet kumar, Ganesh Ramakrishnan and J. Saketha Nath. Entity Resolution and Location Disambiguation in Ancient Hindu Temples Domain using Web Data. Accepted (demo track paper) in NAACL-HLT, 2018..
- Arun Iyer, Saketha Nath J and Sunita Sarawagi. Privacy-preserving Class Ratio Estimation. ACM SIG KDD 2016 (pdf).
- Pratik J., Manik Varma and Saketha Nath J. On p-norm Path Following in Multiple Kernel Learning for Non-linear Feature Selection. ICML 2014 (pdf).
- Arun Iyer, Saketha Nath J and Sunita Sarawagi. Maximum Mean Discrepancy for Class Ratio Estimation: Convergence Bounds and Kernel Selection. Accepted in ICML 2014 (pdf).
- Ankit Ramteke, Akshat Malu, Pushpak Bhattacharyya and Saketha Nath. Detecting Turnarounds in Sentiment Analysis: Thwarting. ACL Short papers 2013.
- Pratik J., and J. Saketha Nath. A Convex Feature Learning Formulation for Latent Task Structure Discovery. ICML-2012, pdf.
- Pratik J., J. Saketha Nath and Ganesh R. Efficient Rule Ensemble Learning using Hierarchical Kernels. ICML-2011. pdf | techrep | code.
- Pratik J., and J. Saketha Nath. Multi-task Multiple Kernel Learning. SDM2011. Draft. Code.
- J. Saketha Nath, G. Dinesh, S. Raman, C. Bhattacharyya, A. Ben-Tal, K. R. Ramakrishnan. On the Algorithmics and Applications of a Mixed-norm based Kernel Learning Formulation. Advances in Neural Information Processing Systems (NIPS), Vancouver, 2009. (pdf)
- S. Bhadra, J. Saketha Nath, A. Ben-Tal and C. Bhattacharyya. Interval Data Classification under Partial Information: A Chance-Constraint Approach. In Proceedings of the PAKDD conference, Bangkok, 2009. [Best Paper - Runner Up]. pdf | slides
- R. Babaria, J. Saketha Nath, S. Krishnan, Sivaramakrishnan, C. Bhattacharyya and M. N. Murty. Focussed Crawling with Scalable Ordinal Regression Solvers. In Proceedings of the ICML conference, Oregon, 2007. pdf | slides | poster
- J. Saketha Nath and C. Bhattacharyya. Maximum Margin Classifiers with Specified False Positive and False Negative Error Rates. In Proceedings of SDM conference, Minneapolis, 2007. pdf
- J. Saketha Nath, C. Bhattacharyya and M. N. Murty. Clustering Based Large Margin Classification: A Scalable Approach using SOCP Formulation. In Proceedings of the SIGKDD conference, Philadelphia, 2006. pdf
Journals
- Karthik Raman, Sowmya Manojna Narasimha, Tanisha Malpani, Omkar Mohite, and Saketha Nath. Understanding flux switching in metabolic networks through an analysis of synthetic lethals. Accepted in Systems Biology and Applications
- Piyushi Manupriya, J. Saketha Nath and Pratik Jawanpuria. MMD-Regularized Unbalanced Optimal Transport. Accepted in TMLR journal, Jan-2024. pdf
- Pratik Jawapuria, J. SakethaNath and Ganesh Ramakrishnan. Generalized Hierarchical Kernel Learning. Journal of Machine Learning Research, vol. 16, Pg. 617-652, Mar 2015 pdf.
- J. Saketha Nath, A. Ben-Tal and C. Bhattacharyya. Robust formulations for clustering-based large-scale classification. Journal of Optimization & Engg., vol. 14(2), Pg. 225-250, June 2013. pdf | code | page1|page2
- J. Aflalo, A. Ben-Tal, C. Bhattacharyya, J. Saketha Nath and S. Raman. Variable Sparsity Kernel Learning. Journal of Machine Learning Research, vol. 12, Pg. 565-592, 2011. (web-page, code, pdf)
- A. Ben-Tal, S. Bhadra, C. Bhattacharyya and J. Saketha Nath. Chance Constrained Uncertain Classification via Robust Optimization. Mathematical Programming Series B (special issue on Machine Learning), vol. 127(1), Pg. 145-173, 2010. (pdf)
- J. Saketha Nath and S. K. Shevade. An efficient clustering scheme using support vector methods. Pattern Recognition, vol. 39(8), Pg. 1473-1480, 2006. pdf | code
Thesis
- Phd Thesis: Learning Algorithms using Chance-Constrained Programs. pdf
- MTech Thesis: An Efficient Clustering Scheme using Support Vector Methods
- BTech Thesis: Scheduling of parts to machines with different capabilities
Professional Activities
- Regular reviewer for NeurIPS, ICML and sometimes for others like AISTATS, AAAI, IJCAI, SDM, JMLR, ICLR etc.
- Guest co-editor for sadhana journal.
Students@IITB
- Past PhD Students: Pratik Jawanpuria (Principal Applied Scientist at Microsoft, Hyderabad) (thesis). Arun Iyer (Principal Applied Scientist at Microsoft Research, Bengaluru) (co-advised with Sunita).
- Past MTech+BTech Students: Devdatta Kathale (MTP), Akshat Jaiswal (co-advised MTP with Sunita), Swati Anand (co-advised MTP with Ganesh), Yugal Sachdev (MTP), Indradyumna Roy, Vivek (co-advised with Siddarth), Laxman Vemula (MTP), Krishna Pillutla (BTP), Rahul Mitra (co-advised MTP), Manasa(MTP), Rakesh (co-advised MTP), Rajesh (MTP), Sarath (MTP), Lokesh (MTP), Srijit (co-advised MTP), Abhinav (co-advised MTP), Harshit Mittal (BTP), Prashanth(BTech), Amit Deoda (DDP), Hemendra (DDP).
Students@IITH
- BTech Students: Lokesh Badisa,
- MTech Students: Anirudh Joshi, Sameer Atram
- PhD Students: Piyushi
- Past Students: C. Geetha Sowmya, K.V.D Sri Harsha, Prabhath Chellangi, P. Ganesh N. Madhav, Sayantan Biswas, Rachit Keerti Das, Nishanth (Thesis (MTP)), Aishwarya Gurjar (MTP), Aditya Saibewar (MTP), Zeeshan Ali (MTP)