Rohan Ghuge

Assistant Professor
IROM, McCombs School of Business
The University of Texas at Austin

CV

About Me

I am an Assistant Professor of Decision Science in the Department of Information, Risk, and Operations Management (IROM) at the McCombs School of Business at the University of Texas at Austin. Prior to joining the IROM department, I was a Ronald J. and Carol T. Beerman / ARC Postdoctoral Fellow at the Georgia Institute of Technology, working with Vidya Muthukumar, Mohit Singh and Sahil Singla. And before that, I completed my Ph.D. in Industrial and Operations Engineering at the University of Michigan where I was advised by Viswanath Nagarajan.

Research Interests

My research focuses on designing data-driven algorithms for problems involving uncertainty. In particular, I employ data-driven methods and machine learning techniques in stochastic optimization, where the input is unknown but the distribution over input instances is known. My work centers on three primary themes:


  • exploring the power of adaptivity—the ability of an algorithm to adjust its decisions based on newly acquired feedback, particularly focusing on how limited rounds of feedback can still yield near-optimal solutions;

  • developing robust algorithms for online decision-making, which involves making irrevocable choices without foresight, with an emphasis on ensuring reliable performance under noisy or uncertain data; and

  • tackling the dual challenge of learning problem parameters while simultaneously optimizing decisions, with a particular focus on analyzing the convergence rate of such algorithms to optimal solutions.

These considerations are critical in applications like healthcare diagnostics, preference elicitation on online platforms, ad placement in response to search queries, and web search ranking. Through this work, I aim to advance both the practical and theoretical understanding of integrating machine learning techniques into decision-making under uncertainty.

Journal Publications

  • Informative Path Planning with Limited Adaptivity.
    Rayen Tan, Rohan Ghuge and Viswanath Nagarajan. [PDF]
    INFORMS Journal of Computing (Minor Revision).
    Full version of AISTATS 2024 paper.

  • Non-Adaptive Stochastic Score Classification annd Explainable Halfspace Evaluation
    Rohan Ghuge, Anupam Gupta and Viswanath Nagarajan. [PDF]
    Operations Research (Articles in Advance). [DOI]
    Full version of IPCO 2022 paper.

  • The Power of Adaptivity for Stochastic Submodular Cover.
    Rohan Ghuge, Anupam Gupta and Viswanath Nagarajan. [PDF]
    Operations Research, 72(3): 1156-1176, 2024. [DOI]
    Full version of ICML 2021 paper.

  • Constrained Assortment Optimization under the Paired Combinatorial Logit Model.
    Rohan Ghuge, Joseph Kwon, Viswanath Nagarajan and Adetee Sharma. [PDF]
    Operations Research, 70(2): 786-804, 2022. [DOI]

  • Quasi-Polynomial Algorithms for Submodular Tree Orienteering and Other Directed Network Design Problems.
    Rohan Ghuge and Viswanath Nagarajan. [PDF]
    Mathematics of Operations Research, 47(2):1612-1630, 2022. [DOI]
    Full version of SODA 2020 paper.


Conference Publications

  • Improved and Oracle-Efficient Online l1 l_1 -Multicalibration.
    Rohan Ghuge, Vidya Muthukumar, and Sahil Singla. [PDF]
    International Conference on Machine Learning (ICML 2025).

  • Single-Sample and Robust Online Resource Allocation.
    Rohan Ghuge, Sahil Singla and Yifan Wang. [PDF]
    Symposium on Theory of Computing (STOC 2025).

  • Semi-Bandit Learning for Monotone Stochastic Optimization.
    Arpit Agarwal, Rohan Ghuge and Viswanath Nagarajan. [PDF]
    Symposium on Foundations of Computer Science (FOCS 2024). [DOI].

  • Informative Path Planning with Limited Adaptivity.
    Rayen Tan, Rohan Ghuge, and Viswanath Nagarajan. [PDF]
    International Conference on Artificial Intelligence and Statistics (AISTATS 2024). [DOI].

  • An Asymptotically Optimal Batched Algorithm for the Dueling Bandit Problem.
    Arpit Agarwal, Rohan Ghuge, and Viswanath Nagarajan. [PDF]
    Neural Information Processing Systems (NeurIPs 2022). [DOI].

  • Batched Dueling Bandits.
    Arpit Agarwal, Rohan Ghuge, and Viswanath Nagarajan. [PDF]
    International Conference on Machine Learning (ICML 2022). Long talk (top 2% of submissions). [DOI].

  • Non-Adaptive Stochastic Score Classification and Explainable Halfspace Evaluation.
    Rohan Ghuge, Anupam Gupta and Viswanath Nagarajan. [PDF]
    International Conference on Integer Programming and Combinatorial Optimization (IPCO 2022). [DOI].

  • The Power of Adaptivity for Stochastic Submodular Cover.
    Rohan Ghuge, Anupam Gupta and Viswanath Nagarajan. [PDF]
    International Conference on Machine Learning (ICML 2021). Long talk (top 3% of submissions). [DOI].

  • Quasi-Polynomial Algorithms for Submodular Tree Orienteering and Other Directed Network Design Problems.
    Rohan Ghuge and Viswanath Nagarajan. [PDF]
    Symposium on Discrete Algorithms (SODA 2020). [DOI]

Teaching Experience

  • At Georgia Institute of Technology, as an Instructor.

    • Summer 2024, Online Learning and Decision-Making (ISyE 4601)
  • At the University of Michigan, as a Graduate Student Instructor.

    • Fall 2019, Stochastic Processes I (IOE 515)
    • Winter 2020, Advanced Optimization Methods (IOE 410)

  • At the University of Pennsylvania, as a Teaching Assistant.

    • Spring 2017, Introduction to Algorithms (CIS 320)
    • Summer 2017, Analysis of Algorithms (CIS 502)
    • Fall 2017, Machine Learning (CIS 520)
    • Spring 2018, Introduction to Algorithms (CIS 320)
    • Spring 2018, Introduction to Probability (ESE 301)

Visit my blog here.