Learning & Optimization
for Autonomous Intelligent
Networked Systems.
I develop mathematical foundations and algorithms with provable guarantees for intelligent systems that learn safely, adapt efficiently, and coordinate under uncertainty.
Department of Electrical and Computer Engineering
University at Buffalo, The State University of New York
Safe cooperative learning with coupled instantaneous hard constraints.
Read the NeurIPS paper → AcceptedAge of Intelligence: decision-aware model freshness and cost-aware updates.
Fresh Enough to Decide → SpotlightNear-optimal adversarial reinforcement learning with switching costs.
Read the spotlight paper → Best Paper Award Runner-UpJoint online provisioning and queue-aware scheduling with switching costs and cross-level constraints.
Read the recognized paper →In the news · UB Engineering: preparing networks for 6G →
Performance and safety guarantees for sequential decisions under uncertainty.
Learning under hard constraints on each action, including when agents interact.
Near-optimal safe reinforcement learning, non-convex safety certification, and coupled-agent learning. The NeurIPS algorithm combines local safety certificates, optimistic value learning, and a known safe fallback; its safety and regret guarantees depend on the paper’s stated assumptions.
Online decisions that weigh performance against switching, reconfiguration, and resource costs.
Competitive online optimization, switching-aware learning, and resource provisioning. In the studied adversarial bandit model with switching costs, allowing two arms to be selected per round improves the achievable dependence of regret on the time horizon T from Θ(T2/3) to O(√T), without full-information feedback.
Understanding the value of partial observations, preference feedback, and fresh models.
Regret bounds and information limits for partially observable learning, alongside decision-aware synchronization that balances model staleness against refresh costs.
My long-term goal is to connect safety, information, and adaptation in one mathematical framework for networked intelligence.
Safe autonomy, dependable digital infrastructure, and efficient edge AI motivate this research. The linked papers provide theoretical foundations and numerical evidence toward these goals.
Our next questions concern safe coordination, selective observation, and model updates across cloud platforms, edge servers, and end devices under changing conditions.
Explore the research program →At UB, I mentor student-led research and connect mathematics to engineering systems through K–12 teacher engagement and undergraduate mentoring. At Ohio State, I supported the NSF AI-EDGE Research Experiences for Undergraduates (REU) program and engaged young learners through LEGO FIRST League.
For prospective students and collaborators in learning theory, optimization, and intelligent systems.
Outside research, I enjoy photography and spending time with my cat, Maomi. A few quiet moments, through my lens.
Select a photograph for a larger view.
Original photographs credited to Ming Shi are © Ming Shi. All rights reserved. For reuse permissions, please contact me. Image reuse policy →