معرفی
Ashutosh Trivedi is an Associate Professor of Computer Science at the University of Colorado Boulder, specializing in formal methods for safety-critical learning-enabled systems. His research bridges computer science, control theory, and machine learning with a focus on trustworthy AI systems that operate safely, fairly, and responsibly.
His research interests encompass Formal Methods, Reinforcement Learning, Software Fairness, and Software Accountability. Trivedi develops mathematical approaches to bring precision to AI system design, using formal languages, automata, and logic to transform vague natural-language instructions into clear specifications. His work emphasizes principled interaction with AI, creating explainable and reliable systems through techniques like reinforcement learning algorithms for cardiac pacemaker design, SAT solvers for grounding LLM outputs, and formal logic for capturing legal obligations.
His recent publications reveal a strong trend toward integrating formal verification with machine learning systems, particularly focusing on fairness testing, safety constraints in reinforcement learning, and neurosymbolic approaches. The research spans applications from legal-critical software to medical devices, with consistent emphasis on accountability and ethical considerations in AI systems.
- 2022 NSF CAREER award
- Liverpool Fellowship
- Distinguished Paper Award at CAV for Regular Reinforcement Learning
- Senior Member of the ACM
- Royal Society Wolfson Visiting Fellowship
- NeuS 2025 Disruptive Idea Award
Trivedi actively mentors PhD students, having supervised multiple successful thesis defenses including Shadi Tasdighi Kalat on multi-agent games, Mateo Perez on formal languages for reinforcement learning, John Komp on pacemaker therapy, Vishnu Murali on functional proofs, and Taylor Dohmen on sequential optimization. His group, the Programming Languages and Verification (CUPLV) at CU Boulder, focuses on making AI systems more trustworthy through mathematical precision and formal verification techniques.
Currently on sabbatical at the University of Liverpool, Trivedi continues to bridge the gap between machine learning and responsible system design, helping AI systems earn the trust we increasingly place in them for critical applications.
