- Academics
-
Research
-
Centres
- E-Health Research Centre (EHRC)
- Machine Intelligence & Robotics CoE (MINRO)
- Centre for IT & Public Policy (CITAPP)
- Cognitive Computing CoE (CCC)
- Centre for Accessibility in the Global South (CAGS)
- COMET Tech Innovation Hub (NM-ICPS)
- IIITB Innovation Centre
- Modular Open-Source Identity Platform (MOSIP)
-
Labs
- Surgical and Assistive Robotics Lab
- Graphics-Visualization-Computing-Lab
- Web Science Lab
- Multimodal Perception Lab
- Software Engineering Lab
- High Density Electronic Systems Lab
- Networking and Communication Lab
- Remote Sensing, GIS and Spatial Computing Lab
- Indian Knowledge System (IKS) Lab
- Smart City Lab
- Ascend Studio
- Radar Sensing Lab
- CSSMP
- Mahabala Ganaka Labs
- Advanced Wireless Communications Lab
- Speech Lab
- Connected Devices and Wearables Lab
- Outreach
- Publications
- Policy
-
Centres
- Placements
- Campus Life
- Media
- People
- About Us

Social Profile Links
Assistant Professor
Education : Ph.D. (Pennsylvania State University)
Aswin Kannan holds a Doctorate from Pennsylvania State Univ., US State (Advisor: Dr. Uday Shanbhag, Dec. 2014), a Masters from Univ. of Illinois at Urbana Champaign, US (May 2010), and a Bachelors from College of Engineering Guindy, India (May 2008). Prior to joining IIIT Bangalore, he worked for Humboldt Universitaet zu Berlin in the Mathematics department as a faculty member (Group Leader, 2021-2024). Even earlier, he had worked for IBM Research (India, 2015-2021), Oracle Analytics (Boston, US, 2014-2015), and Argonne National Labs (Chicago, US, 2010-2012). His research interests fall at the intersection of multiobjective optimization and machine learning. From an applications standpoint, these cover a wide set of domains from energy systems, imaging and retail to portfolio analytics and compliance.
Currently, on the theoretical side -- he is specifically interested in problems related to machine learning, derivative-free optimization, and multi-objective optimization. On the applications-side, his current interests relate to energy systems, business analytics, and industrial/defense safety.
Theory: Mathematical Optimization and Machine Learning.
Applications: Energy Systems, Industrial and Defense Safety, and Business Analytics.
Working
[W1] Z. Li. and A. Kannan, On Hybrid Schemes for Two-Level Problems in Derivative Free Optimization and Machine Learning, In Preparation, 19 pages, March 2025. Fields: Data-Science/Information Systems/Operations Research.
Journals
[P1] E. Tevruez. and A. Kannan, On Approximations and Adaptive Methods for Multiobjective Unit Commitment, Under Review, Operational Research, 32 pages, January 2025.
[P2] P. Dvurechenskii. C. Geiersbach, M. Hintermueller, A. Kannan, and G. Zoettl, A Cournot–Nash Model for a Coupled Hydrogen and Electricity Market, Under Review at the Springer Journal on Energy Systems, 31 pages, October 2024.
[P3] A. Kannan, T. Kreimeier, and A. Walther, On Solving Nonsmooth Retail Portfolio Maximization Problems Using Active Signature Methods, Under Review at the Computational Management Science Journal, 21 pages, July 2024,
[P4] A. Kannan and U. V. Shanbhag, Pseudomonotone Stochastic Variational Inequality Problems: Anal ysis and Optimal Stochastic Approximation Schemes, Computational Optimization and Applications, Volume 74, No. 3, pp. 779- 820, 2019
[P5] A. Kannan, U. V. Shanbhag, and H. M. Kim, Addressing Supply-side Risk in Uncertain Power Markets: Stochastic Nash Models, Scalable Algorithms and Error Analysis, Optimization Methods and Software, Volume 28, No. 5, pp. 1095- 1138, 2013,
[P6] A. Kannan and U. V. Shanbhag, Distributed Computation of Equilibria in Monotone Nash Games via Iterative Regularization Techniques, SIAM Journal of Optimization, Volume 22, No. 4, pp. 1177 1205, 2012,
[P7] A. Kannan, U. V. Shanbhag, and H. M. Kim, Strategic Behavior in Power Markets under Uncertainty, Springer Journal on Energy Systems, Volume 2, No. 2, pp. 115- 141, 2011,
Conferences:
[P8] Z. Li and A. Kannan, Algorithm Switching for Multiobjective Predictions in Renewable Energy Mar kets, Proceedings of the Learning and Intelligent Optimization Conference, June 2024,
[P9] A Kumar, AM Shalghar, H Chauhan, B Ganesan, R Chaudhuri, and A Kannan, Document structure aware Relation Extraction for Semantic Automation, Proceedings of the IKDD/CODS-COMAD, 2024.
[P10] A. Kannan Benefits of Multiobjective Learning in Solar Energy Prediction, Proceedings of the 2nd An nual AAAI (Association for the Advancement of Artificial Intelligence) Workshop on AI to Accelerate Science and Engineering (AI2ASE), Washington-DC, USA, February 2023
[P11] A. M. Shalghar, A. Kumar, B. Ganesan, A. Kannan, A. Parekh, and G. Shobha Document Struc ture aware Relational Graph Convolutional Networks for Ontology Population, Proceedings of the DLG4NLP Workshop, ICLR (International Conference on Learning Representations), April 2022.
[P12] A. Kannan, A. Choudhury, V. Saxena, S. Raje, P. Ram, A. Verma, and Y. Sabharwal HyperASPO: Fusion of Model and Hyper Parameter Optimization for Multi-objective Machine Learning, Proceedings of the IEEE Conference on Big Data, December 2021.
[P13] A. Kannan and U. V. Shanbhag, Distributed Stochastic Optimization Under Imperfect Information, Proceedings of the Conference on Decision and Control (CDC), pp. 400- 405, December 2015.
[P14] A. Kannan and U. V. Shanbhag, The Pseudomonotone Stochastic Variational Inequality Problem: Analytical Statements and Stochastic Extragradient Schemes, Proceedings of the American Conference on Control (ACC), pp. 2930- 2935, June 2014.
[P15] A. Kannan and U. V. Shanbhag, Distributed Iterative Regularization Algorithms for Monotone Nash Games, Proceedings of IEEE Conference on Decision and Control, pp. 1963- 1968, December 2010.
[P16] A. Kannan, S. M. Wild, The Benefit of Deeper Analysis in Simulation Based Groundwater Problems, Proceedings of the XIX International Conference on Computational Methods in Water Resources, June 2012.
[P17] A. Kannan and V. M. Zavala, A Game-Theoretical Model Predictive Framework for Electricity Markets, Proceedings of Allerton Conference on Communication and Control, pp. 1280- 1285, September 2011.
[P18] A. Kannan and E. Natarajan, AFT of Bio-Fuels used as Alternative Sources of Energy in Engines., Proceedings of COBEM, 19th International Congress of Mechanical Engineering, November 2007
Preprints:
[P19] S. Bandyopadhyay, H. Kara, A. Kannan, and M. N. Murthy, FSCNMF: Fusing Structure and Content via Non-negative Matrix Factorization for Information Network Representation, IBM Preprint, Decem ber 2018, arXiv preprint arXiv:1804.05313v2, [Featured in the list of Popular Embedding Techniques- Link].
[P20] A. Kannan and S. M. Wild, Obtaining Quadratic Models of Noisy Functions, Tech. Report, Argonne National Labs, Lemont, IL, Preprint ANL/MCS-P1975-0912 (25 pages), September 2012, [Includes Solver- Link].
Patents- USPTO (Chronological Order)
[PT1] Multi-Objective Automated Machine Learning (US20220180146A1), Filed as High Priority Patent in December 2020, Granted August 2024, IBM Corporation (6 Authors, Secondary Author).
[PT2] PPO: A mathematical programming approach to solve the promotions planning and optimization problem, US Patent 10,528,903, Granted, Oracle Corporation (3 Authors, Lead Author, Filed Nov. 2015, Granted Jan. 2020, Includes Commercial Solver).
[PT3] Computerized Promotion and Markdown Price Scheduling, US Patent 10,776,803, Granted, Oracle Corporation (2 Authors, Lead Author, Filed March 2016, Granted Sep. 2020).
[PT4] Elevator Movement Plan Generation, US Patent 11097921, Filed April 2018, Granted August 2021, IBM Corporation (4 Authors, Secondary Author).
[PT5] Extracting product drag effect from transaction records , US Patent 11416877B2, Filed August 2017, Granted Aug. 2022, IBM Corporation (9 Authors, Lead Author).
[PT6] Bundling items based on cost and value (US20190362407A1), Filed May 2018, IBM Corporation (5 Authors, Primary Author).
[PT7] Determining Collaborative Enterprise Decisions Based on Regulatory Impacts (US20200380444A1), Filed May 2019, IBM Corporation (3 Authors, Lead Author).
[PT8] Classifying data from de-identified content (US20220343151A1), Filed December 2020, IBM Corpora tion (3 Authors, Primary Author).
[PT9] Multi-objective machine learning with model and hyperparameter optimization fusion (US20230069913A1), Filed September 2021, IBM Corporation (7 Authors, Lead Author).
[PT10] Multi-Objective Optimization of Machine Learning Pipelines (US20230281464A1), Filed, March 2022, IBM Corporation (3 Authors, Primary Author).
Defensive Patents (Chronological Order)
[D1] System and Method to Estimate the Level of Usage of Facilities inside A Room by Participants to Design “‘Meeting Room Avatars” and “Reconfigurable Meeting Rooms”, Defensive Patent (ip.com), Published July 2018, IBM Corporation (4 Authors, Secondary Author).
[D2] A Regulatory Product Design Framework for Efficient Cloud Penetration, Defensive Patent (ip.com), June 2019, IBM Corporation (4 Authors, Lead Author).
[D3] System and Method for Cultural Adaption of Educational Content in a Live Setting, Defensive Patent (ip.com), Published August 2018, IBM Corporation (7 Authors, Secondary Author).
[D4] A Method to Improve Efficacy of Marketing Spend by Inferring Incremental Drag Effect among Prod ucts, Defensive Patent (ip.com), July 2019, IBM Corporation (7 Authors, Primary Author).
[D5] Deploying Structure of Personal Data to Improve Entity Classification, Defensive Patent (ip.com), December 2020, IBM Corporation (3 Authors, Primary Author).
AIM 102: Statistical Machine Learning (Spring 2025, Bachelors, 2nd Sem)- IIIT Bangalore. Co-teaching with Prof. Vishwanath.
AIM 846: Multiobjective Machine Learning (Spring 2025, Masters)- IIIT Bangalore.
MATH 3314440: Derivative Free Optimization, Summer 2024 (Masters).
MATH 3314419: Multiobjective Optimization, Fall 2023 (Bachelors).
MATH 3314430: Multiobjective Machine Learning, Fall 2022 (Masters+Bachelors).
Previous Grants (Ongoing):
(1) Using Mathematical programming to enhance Multiobjective Learning (Mathplus, Germany, AA4-11, Main PI), Granted, February 2022: One PhD student supervision.
Description: We explore the deployment of joint hyperpa rameter and model parameter optimization to improve the quality of Pareto frontiers. The underlying problem aims at finding good machine learning models that fit data from different perspectives, not just accuracy.
(2) Multimodal energy markets (Mathplus, AA4-13. Co-PI), Granted, September 2022: One PhD stu dent supervision.
Description: We investigate joint hydrogen and power markets from a game theoretical perspec tive. This involves gas flow constraints (PDEs), conversion aspects, and power generation/allocation (transmission constraints). We focus on developing algorithms to solve the resulting quasi-variational inequality (qVI).
(1) Member of Invention Development Team at IBM (2021).
(2) Certified Invention Evaluator (Bronze Level Badge, September 2021).
(3) IBM Plateau Award for Invention Innovation- Level II (August 2021) and Level I (June 2019).
(4) IBM High Priority Patent Awards (Search-1), September 2020.
(5) COSP(Triennial Conf. on Stochastic Program.) Award Finalist, August 2010, Halifax.
Current:
(1) Kausthubh Manda (IIITB, Masters, 2025 - ).
(2) Zijun Li (PhD, ongoing, Humboldt Univ., 2022 - ).
Previous (Humboldt):
(1) Ece Tevruez (Graduated, September 2024, Masters) - now at Nebenan.de.
(2) Six Bachelors theses - for more details, check my previous webpage at Humboldt.