Theja Tulabandhula, Ph.D.

Interests

I am broadly interested in designing and implementing machine learning, artificial intelligence and sequential decision making methods in consumer facing applications. I have worked on: (a) supervised learning in the presence of operational information, (b) transportation problems including scheduling and fair ride-sharing, (c) assortment optimization and consumer behavior modeling, (d) new product pricing and learning consumer utility functions, (e) recommendation system design with privacy and regret guarantees, and (f) several online and reinforcement learning problems.

Background

• I am currently an Assistant Professor at the University of Illinois at Chicago.
• I was a Research Scientist at Xerox Research (Machine Learning and Statistics Group, Data Analytics Lab) during 2015-2016.
• I received a PhD in Electrical Engineering and Computer Science from MIT, and was advised by Cynthia Rudin. While there, I was also a Fulbright Science and Technology Scholar.
• Before that, I was an undergrad at IIT Kharagpur, where I was also awarded the Prime Minister's Gold Medal.

Publications and Patents

In addition to below, you can also find my publications at Arxiv or DBLP or SSRN. For many of these works, source code is made available on GitHub. Further, a few of my patents (4+ granted) are listed on Google Patents.

• Mehrnaz Amjadi, Vijay Kamble, and Theja Tulabandhula. Managing Adoption under Network Effects. Proceedings of the 14th Workshop on the Economics of Networks, Systems and Computation: NetEcon, 2019.
• Yunjuan Wang and Theja Tulabandhula. Thompson Sampling for a Fatigue-aware Online Recommendation System. Proceedings of the 4th Multidisciplinary Conference on Reinforcement Learning and Decision Making, 2019.
• Priyank Agrawal and Theja Tulabandhula. Bandits with Temporal Stochastic Constraints. Proceedings of the 4th Multidisciplinary Conference on Reinforcement Learning and Decision Making, 2019.
• Ragavendran Gopalakrishnan, Theja Tulabandhula, and Koyel Mukherjee. Sequential Individual Rationality in Dynamic Ridesharing. SSRN Preprint, 2018.
• Debjyoti Saharoy and Theja Tulabandhula. Learning Buyer Behavior under Realistic Pricing Restrictions. Proceedings of the International Symposium on Artificial Intelligence and Mathematics, 2018. Note - also presented at the INFORMS Revenue Management and Pricing Conference 2018.
• Arpita Biswas, Ragavendran Gopalakrishnan, Theja Tulabandhula, Koyel Mukherjee, Asmita Metrewal, and Raja Subramaniam Thangaraj. Impact of Detour-Aware Policies on Maximizing Profit in Ridesharing. Transportation Research Board 97th Annual Meeting, 2018. Note - also presented at the 10th International Workshop on Agents in Traffic and Transportation 2018 held in conjunction with ECAI/IJCAI, AAMAS and ICML conferences.
• Mehrnaz Amjadi and Theja Tulabandhula. Block-Structure Based Time-Series Models For Graph Sequences. NIPS Workshop on Advances in Modeling and Learning Interactions from Complex Data, 2017. video. Note - also presented as a poster at the 2nd Midwest Machine Learning Symposium 2018.
• Deeksha Sinha and Theja Tulabandhula. Optimizing Revenue over Data-driven Assortments. NIPS Workshop on Nearest Neighbor Methods, 2017.
• U.N. Niranjan, Arun Rajkumar, and Theja Tulabandhula. Provable Inductive Robust PCA via Iterative Hard Thresholding. 33rd Conference on Uncertainty in Artificial Intelligence, 2017.
• Anuj Mahajan and Theja Tulabandhula. Symmetry Detection and Exploitation for Function Approximation in Deep Reinforcement Learning. The 3rd Multidisciplinary Conference on Reinforcement Learning and Decision Making, 2017.
• Sudeep Raja Putta and Theja Tulabandhula. Efficient Reinforcement Learning via Initial Pure Exploration. The 3rd Multidisciplinary Conference on Reinforcement Learning and Decision Making, 2017.
• Vikas Jain and Theja Tulabandhula. Faster Reinforcement Learning Using Active Simulators. NIPS Workshop: Teaching Machines, Robots, and Humans, 2017. Note - also a poster at the 3rd Multidisciplinary Conference on Reinforcement Learning and Decision Making 2017.
• Ragavendran Gopalakrishnan, Koyel Mukherjee, and Theja Tulabandhula. The Costs and Benefits of Ridesharing: Sequential Individual Rationality and Sequential Fairness. INFORMS Transportation and Logistics Society Conference, 2017. Note - also a poster at the 17th ACM Conference on Economics and Computation 2016.
• Arun Rajkumar, Koyel Mukherjee, and Theja Tulabandhula. Learning to Partition Using Pairwise Compatibilities. International Conference on Autonomous Agents and Multiagent Systems, 2017.
• Anuj Mahajan and Theja Tulabandhula. Symmetry Detection and Exploitation for Function Approximation in Deep Reinforcement Learning. International Conference on Autonomous Agents and Multiagent Systems, 2017. Note - superceded.
• Sudeep Raja Putta and Theja Tulabandhula. Pure Exploration in Episodic Fixed-Horizon Markov Decision Processes. International Conference on Autonomous Agents and Multiagent Systems, 2017. Note - superceded.
• Arpita Biswas, Ragavendran Gopalakrishnan, Theja Tulabandhula, Koyel Mukherjee, Asmita Metrewal, and Raja Subramaniam Thangaraj. Profit Optimization in Commercial Ridesharing. International Conference on Autonomous Agents and Multiagent Systems, 2017. Note - superceded.
• Theja Tulabandhula. Interactions Between Learning and Decision Making. AI Matters 3(1), 2017.
• Theja Tulabandhula, Shailesh Vaya, and Aritra Dhar. Privacy-preserving Targeted Advertising. INFORMS Workshop on Data Mining and Decision Analytics, 2016.
• Theja Tulabandhula and Narendra Annamaneni. Optimal Automated Booking of On-demand Transportation in Multi-modal Journeys. Twenty-third World Congress on Intelligent Transportation Systems, 2016.
• Narayanan U. Edakunni, Koyel Mukherjee, Theja Tulabandhula, Kaushik Baruah, Tuhin Bhattacharya, and Geetha Manjunath. Demand Sensitive Scheduling of Public Transport using Past Ticketing Data. Twenty-third World Congress on Intelligent Transportation Systems, 2016.
• Theja Tulabandhula, Koyel Mukherjee, and Narayanan U. Edakunni. Static and Dynamic Scheduling to Minimize Passenger Waiting Times. Transportation Research Board 95th Annual Meeting, 2016.
• Prabuchandran K.J., Tejas Bodas, and Theja Tulabandhula. Reinforcement Learning Algorithms for Regret Minimization in Structured Markov Decision Processes. International Conference on Autonomous Agents and Multiagent Systems, 2016.
• Theja Tulabandhula. Learning Personalized Optimal Control for Repeatedly Operated Systems. NIPS workshop on Machine Learning From and For Adaptive User Technologies: From Active Learning and Experimentation to Optimization and Personalization, 2015.
• Theja Tulabandhula and Cynthia Rudin. On Combining Machine Learning with Decision Making. Machine Learning Journal 97(1-2), 2014.
• Theja Tulabandhula and Cynthia Rudin. Robust Optimization using Machine Learning for Uncertainty Sets. ArXiv Preprint, 2014.
• Theja Tulabandhula and Cynthia Rudin. Robust Optimization using Machine Learning for Uncertainty Sets. Proceedings of the International Symposium on Artificial Intelligence and Mathematics, 2014. Note - superceded by an arxiv preprint.
• Theja Tulabandhula and Cynthia Rudin. Generalization Bounds for Learning with Linear, Polygonal, Quadratic and Conic Side Knowledge. Machine Learning Journal, 2014.
• Theja Tulabandhula and Cynthia Rudin. Generalization Bounds for Learning with Linear and Quadratic Side Knowledge. Proceedings of the International Symposium on Artificial Intelligence and Mathematics, 2014.
• Theja Tulabandhula and Cynthia Rudin. Tire Changes, Fresh Air, and Yellow Flags: Challenges in Predictive Analytics for Professional Racing. Big Data 2(2), 2014.
• Theja Tulabandhula. Interactions Between Learning and Decision Making. Ph.D. thesis, Massachusetts Institute of Technology, 2014.
• Theja Tulabandhula and Cynthia Rudin. Machine Learning with Operational Costs. Journal of Machine Learning Research, 2013.
• Theja Tulabandhula and Cynthia Rudin. The Influence of Operational Cost on Estimation. Proceedings of the International Symposium on Artificial Intelligence and Mathematics, 2012. Note - superceded by a journal version..
• Theja Tulabandhula, Cynthia Rudin, and Patrick Jaillet. The Machine Learning and Traveling Repairman Problem. Proceedings of the Second International Conference on Algorithmic Decision Theory, 2011. Note - superceded by a journal version..
• Theja Tulabandhula. Some Architectures for Chebyshev Interpolation. ArXiv Preprint, 2010.
• Theja Tulabandhula and Yujendra Mitikiri. A 20MS/s 5.6 mW 6b Asynchronous ADC in 0.6$\mu$m CMOS. Proceedings of the 22nd International Conference on VLSI Design, 2009.
• Theja Tulabandhula, Samuel Antao, and Leonel Sousa. A Class of Software-Hardware Processors for Fingerprint Matching on the Fourier Domain. Proceedings of the Third HiPEAC Workshop on Reconfigurable Computing, 2009.
• Theja Tulabandhula, Amit Patra, and Nirmal B Chakrabarti. Design of a Two Dimensional PRSI Image Processor. Proceedings of the 11th EUROMICRO Conference on Digital System Design: Architectures, Methods and Tools, 2008.

Social Media

• Twitter
• Linkedin
• Github
• Email: work AT domain

Teaching

• Statistical Models and Methods (Fall 2019)
• Advanced Predictive Models (Fall 2019)
• Statistical Models and Methods (Spring 2019): notes, resources and videos available!
• Advanced Predictive Models (Spring 2019): notes, resources and videos available!
• Statistical Models and Methods (Spring 2018): notes and videos available!
• Advanced Predictive Models (Spring 2018): slides and videos available!
• Statistical Models and Methods (Fall 2017): videos available!
• Advanced Predictive Models (Spring 2017): videos available!