Hi, I am Shourya Bose. I am a PhD candidate with the eODAL Lab at University of California, Santa Cruz. My advisor is Dr. Yu Zhang. My research interests span the intersection of machine learning, optimization, and control theory, all applied to electric power grids. Within machine learning, my works deal with reinforcement learning, convex optimization accelerators, federated learning, and time-series foundation models.

Previously I was a Research Associate (RA) at the Department of Electrical Engineering at Indian Institute of Science (IISc) Bengaluru in the Control and Network Systems Group.

I graduated with a Dual Degree from BITS Pilani (KK Birla Goa Campus) on December, 2019. I majored in M.Sc. (Hons.) Mathematics, and B.E. (Hons.) Electrical and Electronics Engineering. I graduated with a First Class grade in both the subjects.

You can download my CV here.

Jump to section: Publications, Other Works, Miscellaneous

First Author Publications - Journal

Federated Short-Term Load Forecasting with Personalization Layers for Heterogeneous Clients (arXiv Link)

Authors: Shourya Bose, Kibaek Kim. Report.

A study on applications of specialized mixture-of-experts models for accelerating convex optimal power flow problems

Authors: Shourya Bose, Kejun Chen, Yu Zhang. Exact title and manuscript unavailable due to double-blind review requirements.

Load Restoration in Islanded Microgrids: Formulation and Solution Strategies (Link)

Authors: Shourya Bose, Yu Zhang. IEEE Transactions on Control of Network Systems, 2023. Received INFORMS ENRE Early Career Best Paper Award.

Event-Triggered Second-Moment Stabilisation under Action-Dependent Markov Packet Drops (Link)

Authors: Shourya Bose, Pavankumar Tallapragada, IET Control Theory and Applications, 2021.

First Author Publications - Conference

From RNNs to Foundation Models: An Empirical Study on Commercial Building Energy Consumption (Link)

Authors: Shourya Bose, Yijiang Li, Amy Van Sant, Yu Zhang, Kibaek Kim

Addressing Heterogeneity in Federated Load Forecasting with Personalization Layers (Link)

Authors: Shourya Bose, Yu Zhang, Kibaek Kim. IISE Annual Conference and Expo, 2024.

Privacy-Preserving Load Forecasting for Personalized Model Obfuscation (Link)

Authors: Shourya Bose, Yu Zhang, Kibaek Kim. IEEE PES-GM 2024.

On LinDistFlow Model Congestion Pricing: Bounding the Changes in Power Tariffs (Link)

Authors: Shourya Bose, Kejun Chen, Yu Zhang. IEEE ISGT 2023. Received IEEE SYPA Travel Grand, UCSC Dean’s Travel Grant.

Co-optimization of Battery Routing and Load Restoration for Microgrids with Mobile Energy Storage Systems (Link)

Authors: Shourya Bose, Sifat Chowdhury, Yu Zhang, IEEE PES-GM 2022.

Differentially Private Load Restoration for Microgrids with Distributed Energy Storage (Link)

Authors: Shourya Bose, Yu Zhang, IEEE ISGT NA 2022.

Event-Triggered Second Moment Stabilization under Markov Packet Drops (Link)

Authors: Shourya Bose, Pavankumar Tallapragada. Fifth Indian Control Conference, 2019.

Other Works

Pixxel Space

I worked as a research intern at Pixxel Space. Pixxel is aiming to launch a constellation of earth observing satellites and revolutionize space-based data collection by using it to solve day-to-day problems.

As a part of the company, I was involved in spacecraft orbit simulation.

I also gained exposure to softwares such as NASA’s GMAT, AGI’s STK, and a.i. Solutions’ FreeFlyer.

Miscellaneous

Learning to Run a Power Grid (L2RPN, 2023)

Out team (including myself, Qiuling Yang, and Yu Zhang) were the winners of the L2RPN 2023 competition organized by TU Delft, which saw 95 teams from across the world compete in a competition to generate reinforcement-learning based methods to operate a power grid. The control action involved reconfiguring the grid topology, controlling the outputs of generators, and isolating various buses. Our strategy involved generating atomic actions which, when applied, reduced the control task to a continuous optimization problem which can be easily solved.

Hult Prize 2016 (Dubai Finals)

Hult Prize is an annual entrepreneurship competition organized on a global scale to encourage innovative entrepreneurial solutions to large-scale social problems. I was a part of Project Vriddhi, a team of 4 BITS Pilani, Goa students. We proposed an idea for urban decongestion using scheduled & shared transportation. This was in response to solve the problem of urban crowded slums, which was the Hult Prize Challenge for 2016.