Ashutosh Trivedi

Ashutosh Trivedi

Associate Professor of Computer Science · University of Colorado Boulder

I work on formal methods for reinforcement learning, trustworthy AI, and safety-critical software and cyber-physical systems.

CV Google Scholar DBLP Email

Research

Formal Methods for Reinforcement Learning

Verification, synthesis, and symbolic reasoning for reinforcement-learning systems, including temporal objectives, recursion, and structured representations.

Trustworthy Reasoning with LLMs

Formal and symbolic methods for producing explanations and guarantees with rigorous foundations.

AI, Software, and Accountability

Auditing, testing, and debugging methods for high-stakes software, with applications to fairness, legal compliance, and socio-technical decision-making.

Secure & Safe Cyber-Physical Systems

Security, privacy, and safety methods for cyber-physical and learning-enabled systems, with applications to medical devices and critical infrastructure.

Recent news

Feb 2026
Hiring two Student Research Assistants for a project on Cardiac Digital Twins and Reinforcement Learning (Spring 2026). Details here.
Aug 2025
Teaching CSCI 5444 (Introduction to the Theory of Computation) this term.
Aug 2025
Our research on AI explainability in Sudoku was featured in CNET.
Aug 2025
New arXiv preprint with Maria Leonor Pacheco, Fabio Somenzi, and Dananjay Srinivas: Explaining Hitori Puzzles: Neurosymbolic Proof Staging for Sequential Decisions.
Aug 2025
Paper accepted at ASE 2025: Uncovering Discrimination Clusters: Quantifying and Explaining Systematic Fairness Violations.

All news →

Selected publications

Stochastic Neural Simulation Relations for Control Transfer

🏆 DARPA Disruptive Idea Award
cps award
NeuS · 2025
PDF

Regular Reinforcement Learning

🏆 CAV Distinguished Paper Award
award formalrl trustworthyAI
CAV · 2024
PDF Details

Closure Certificates

cps theory
HSCC 2024 · 2024
PDF Details

Metamorphic Testing and Debugging of Tax Preparation Software

accountableSE trustworthyAI
ICSE-SEIS 2023 · 2023
PDF Details

Recursive Reinforcement Learning

formalrl theory trustworthyAI
NeurIPS · 2022
PDF

Omega-Regular Objectives in Model-Free Reinforcement Learning

formalrl theory trustuworthyAI
TACAS · 2019
PDF Details