Publications (Google Scholar) (LinkedIn)

Dissertation

• On the Value of Prediction and Feedback for Online Decision Making With Switching Costs
  

  Ming Shi
  

  Purdue University Doctoral Dissertation, June 2022.

Book Chapter

• Adversarial Online Reinforcement Learning Under Limited Defender Resources
  

  Ming Shi, Yingbin Liang, and Ness B. Shroff
  

  Network Security Empowered by Artificial Intelligence, Springer, Series ISSN 1568-2633, June 2024.

Journal Papers

• Power-of-2-Arms for Adversarial Bandit Learning with Switching Costs
  

  Ming Shi, Xiaojun Lin, and Lei Jiao
  

  IEEE/ACM Transactions on Networking, vol. 33, no. 3, pp. 1112-1127, June 2025, DOI: 10.1109/TON.2024.3522073. [IEEE/ACM ToN]

• Combining Regularization With Look-Ahead for Competitive Online Convex Optimization
  

  Ming Shi, Xiaojun Lin, and Lei Jiao
  

  IEEE/ACM Transactions on Networking, vol. 32, no. 3, pp. 2391-2405, June 2024, DOI: 10.1109/TNET.2024.3350990. [IEEE/ACM ToN]

• Competitive Online Convex Optimization with Switching Costs and Ramp Constraints
  

  Ming Shi, Xiaojun Lin, and Sonia Fahmy
  

  IEEE/ACM Transactions on Networking, vol. 29, no. 2, pp. 876-889, April 2021, DOI: 10.1109/TNET.2021.3053910. [IEEE/ACM ToN]

• Near-Optimal Partially Observable Reinforcement Learning with Partial Online State Information
  

  Ming Shi, Yingbin Liang, and Ness B. Shroff
  

  submitted to IEEE Transactions on Information Theory, 2025. [IEEE TIT]

• Theoretical Guarantees for Reinforcement Learning with Noisy Multi-Human Feedback
  

  Ming Shi, Yingbin Liang, Ness B. Shroff, and Ananthram Swami
  

  submitted to Journal of Machine Learning Research, January 2026. [JMLR]

• Physics-Informed Stochastic Post-Decision State Reinforcement Learning for Computationally-Efficient IoT Sensor Scheduling
  

  Alireza Mohammadhosseini, Andrew Corra, Ming Shi, Nicholas Mastronarde, and Jacob Chakareski
  

  submitted to IEEE Journal of Selected Areas in Sensors [IEEE JSAS]

Conference Papers (* My students are marked by "*...*" *)

• Provably Efficient Personalized Multi-Objective Bandits with Proactive Conversational Queries
  

  Linfeng Cao, Ming Shi, and Ness B. Shroff
  

  submitted to 42nd Conference on Uncertainty in Artificial Intelligence. [UAI]

• Stability Limits of Diffusion Robustness Under Reverse-Process Errors
  

  Ming Shi
  

  submitted to 39th Annual Conference on Learning Theory. [COLT]

• One Model for All Preferences: Meta-Learning for Multi-Objective LLM Alignment
  

  Fatemeh Nourzad, Daouda Sow, Yingbin Liang, Ming Shi, Ming Zhang, Yunxuan Li, Eylem Ekici, Ness Shroff
  

  submitted to 43rd International Conference on Machine Learning. [ICML]

• Probe-then-Commit Multi-Objective Bandits: Theoretical Benefits of Limited Multi-Arm Feedback
  

  Ming Shi
  

  submitted to 24th International Symposium on Modeling and Optimization in Mobile, Ad hoc, and Wireless Networks. [WiOPT]

• Bi-Level Online Provisioning and Scheduling with Switching Costs and Cross-Level Constraints
  

  *Jialei Liu*, C. Emre Koksal, Ming Shi
  

  submitted to 24th International Symposium on Modeling and Optimization in Mobile, Ad hoc, and Wireless Networks. [WiOPT]

• Barriers to Pareto Steerability in Preference-Conditioned LLM Alignment
  

  Fatemeh Nourzad, Daouda Sow, Yingbin Liang, Ming Shi, Ming Zhang, Yunxuan Li, Eylem Ekici, Ness Shroff
  

  14th International Conference on Learning Representations (I Can't Believe It's Not Better: Challenges in Applied Deep Learning), Rio de Janeiro, Brazil, April 2026. [ICLR-ICBINB]

• One Model, Many Goals: Meta-Learning Preference-Conditioned Alignment for Lifelong LLM Agents
  

  Fatemeh Nourzad, Daouda Sow, Yingbin Liang, Ming Shi, Ming Zhang, Yunxuan Li, Eylem Ekici, Ness Shroff
  

  14th International Conference on Learning Representations (Lifelong Agents: Learning, Aligning, Evolving), Rio de Janeiro, Brazil, April 2026. [ICLR-LLA]

• Minimax Optimal Adversarial Reinforcement Learning
  

  Yudan Wang, Kaiyi Ji, Ming Shi, Shaofeng Zou
  

  14th International Conference on Learning Representations, Rio de Janeiro, Brazil, April 2026. [ICLR] (Acceptance rate: 28%.)

• Communication–Corruption Coupling and Verification in Cooperative Multi-Objective Bandits
  

  Ming Shi
  

  submitted to IEEE International Symposium on Information Theory, January 2026. [IEEE ISIT]

• Mixture-of-Experts Actor-Critic for Regime-Switching MDPs: Impossibility Results and Performance Guarantees
  

  Ming Shi
  

  submitted to ACM SIGMETRICS, January 2026. [ACM SIGMETRICS]

• A Queueing-Theoretic Framework for Dynamic Attack Surfaces: Data-Integrated Risk Analysis and Adaptive Defense
  

  Jihyeon Yun, Abdullah Etcibasi, Ming Shi, and C. Emre Koksal
  

  submitted to ACM SIGMETRICS, January 2026. [ACM SIGMETRICS]

• Provably Efficient Multi-Objective Bandit Algorithms under Preference-Centric Customization
  

  Linfeng Cao, Ming Shi, and Ness B. Shroff
  

  The 40th Annual AAAI Conference on Artificial Intelligence, Singapore, January 2026. [AAAI] (Acceptance rate: 17.6%.)

• Online Learning for Optimizing AoI-Energy Tradeoff under Unknown Channel Statistics
  

  Mohamed A. Abd-Elmagid, Ming Shi, Eylem Ekici, and Ness B. Shroff
  

  26rd International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, October 2025. [ACM MobiHoc] (Acceptance rate: 23%.)

• Provably Efficient Reinforcement Learning for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces
  

  Amirhossein Roknilamouki, Arnob Ghosh, Ming Shi, Fatemeh Nourzad, Eylem Ekici, and Ness B. Shroff
  

  42th International Conference on Machine Learning, Vancouver, Canada, July 2025. [ICML] (Acceptance rate: 26.9%.)

• Designing Near-Optimal Partially Observable Reinforcement Learning
  

  Ming Shi, Yingbin Liang, and Ness Shroff
  

  IEEE Military Communications Conference, Washington DC, October 2024. [IEEE MILCOM]

• A Near-Optimal Algorithm for Safe Reinforcement Learning Under Instantaneous Hard Constraints
  

  Ming Shi, Yingbin Liang, and Ness Shroff
  

  40th International Conference on Machine Learning, Hawaii, USA, July 2023. [ICML] (Acceptance rate: 27.96%.)

• Near-Optimal Adversarial Reinforcement Learning with Switching Costs
  

  Ming Shi, Yingbin Liang, and Ness Shroff
  

  11th International Conference on Learning Representations, Kigali, Rwanda, May 2023. [ICLR] (Spotlight, with acceptance rate 8.0%.)

• Leveraging Synergies Between AI and Networking to Build Next Generation Edge Networks
  

  Ming Shi, Sen Lin (co-first author), Yingbin Liang, Ness Shroff, et al.
  

  8th IEEE International Conference on Collaboration and Internet Computing, virtual conference, December 2022. [IEEE CIC]

• Power-of-2-Arms for Bandit Learning with Switching Costs
  

  Ming Shi, Xiaojun Lin, and Lei Jiao
  

  23rd International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing, Seoul, South Korea, October 2022. [ACM MobiHoc] (Acceptance rate: 19.8%.)

• Combining Regularization with Look-Ahead for Competitive Online Convex Optimization
  

  Ming Shi, Xiaojun Lin, and Lei Jiao
  

  IEEE Conference on Computer Communications, virtual conference, May 2021. [IEEE INFOCOM] (Acceptance rate: 19.9%.)

• Competitive Online Convex Optimization with Switching Costs and Ramp Constraints
  

  Ming Shi, Xiaojun Lin, Sonia Fahmy, and DongHoon Shin
  

  IEEE Conference on Computer Communications, Honolulu, Hawaii, USA, April 2018. [IEEE INFOCOM] (Acceptance rate: 19.2%.)

Technical Reports

• Power-of-2-Arms for Bandit Learning with Switching Costs
  

  Ming Shi, Xiaojun Lin, and Lei Jiao, October 2022.
  

• Combining Regularization with Look-Ahead for Competitive Online Convex Optimization
  

  Ming Shi, Xiaojun Lin, and Lei Jiao, May 2021.
  

• Competitive Online Convex Optimization with Switching Costs and Ramp Constraints
  

  Ming Shi, Xiaojun Lin and Sonia Fahmy, April 2021.