news

- Our work Discrete-Guided Diffusion for Scalable and Safe Multi-Robot Motion Planning was accepted at AAAI 2026! Congrats to Jinhao!
- Check out our new work on accelerating LLM inference with discrete diffusion models! These results are amazing!
- I gave a talk at the ORNL AI Core workshop on Generative AI for Science.

- Check out our new work on manifold constrained protein design with diffusion models!.
- Several new preprints on topics spanning from physics-constrained generative flow matching and diffusion models to safety in agentic AI systems and learning to solve PDE-constrained problems. Check out our publications’ page for details.

- Two papers on physics-constrained generative diffusion models (as Spotlight) and logic-constrained language diffusion models accepted at NeurIPS 2025 Congratulations to Jacob, Michael, and Stefano!
- Our book Differential Privacy in Artificial Intelligence: From Theory to Practice is now online, available as Open Access!
This book covers the (i) theoretical underpinnings of Differential Privacy, (ii) recent algorithmic advances for machine learning,, (iii) practical applications across key engineering and scientific domains, (iv) methodologies for implementation and empirical evaluation, and (v) the surrounding legal and ethical frameworks.
The book is also available on Amazon and Barnes & Noble.

- New preprint: Discrete-Guided Diffusion for Scalable and Safe Multi-Robot Motion Planning. We provide a new integration of discrete optimization and diffusion models for multi-robot motion planning scaling to 100 of agents in cluttered enviroments.
- New preprint: Stability-Constrained AC Optimal Power Flow–A Gaussian Process-Based Approach. We propose a new, scalable approach to incorporate generator dynamics into the ACOPF using Gaussian Process (GP) models. Crucially, this enables probabilistic stability assessment to be integrated directly into the optimization process!
- NSF-RI proposal on Generative Models for Scientific Exploration funded! We’ll integrate optimization and physical principles dirctly in generative models for a variety of scientific applications! Thank you, NSF!
- Paper Privacy-Preserving Convex Optimization: When Differential Privacy Meets Stochastic Programming accepted at CDC-2025!
- New preprint: SoK: Data Minimization in Machine Learning. We provide introduces a comprehensive framework for data minimization in machine learning and introduce a systematical review the literature on data minimization and adjacent methodologies.

- Congratulations to Michael Cardei for winning an NSF GRFP Scholarship Award!
- I gave a talk at USNCCM18 about our generative AI for material discovery.

- New preprint: Disclosure Audits for LLM Agents. We provide a model to audit disclosure risks for LLM agentic systems with conversational interfaces.
- Our paper on Neuro-symbolic Generative Diffusion was awarded a DARPA Disruptive Idea Award! Thank you, DARPA for the support!
- We received two 4-VA research grants! We will work with George Mason and Virginia Tech to build more aligned and interpretable LLM models and more robust learning to optimize pipelines. Thank you 4-VA for your support!

- Grateful to receive research credits from Google Cloud! We will develop more secure agentic LLM systems . Thank you, Google!
- Multi-Agent Path Finding in Continuous Spaces with Projected Diffusion Models was accepted at ICML 2025. Congraulations to Jinhao and Jacob!
- Neuro-symbolic Generative Diffusion Models for Physically Grounded, Robust, and Safe Generation was accepted at Neu 2025.
Congratulations to Jacob, Michael, and Jinhao!
- Congratulations to Joon Ko, who was awarded an Excellence in Research Award for undergraduate research at UVA CS!

- Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion was selected for an Oral presentation to NAACL 2025. Congraulations to Jacob and Michael!
- Our paper The Data Minimization Principle in Machine Learning was accepted at FAccT 2025. Congratulations to Prakhar!
- Congratulations to Michael Cardei, who was awarded an NSF GRFP Honorable Mentions!

- Greatful to receive an NVIDIA Academic Grant Program Award! We will develop faster LLM inference. Thank you, NVIDIA!
- Excited to share our lab’s new work: Constrained Language Generation with Discrete Diffusion Models!
- Our paper Low-rank finetuning for LLMs: A fairness perspective was awarded a Best paper award at the AAAI 2025 colorai workshop.

- Learning To Solve Differential Equation Constrained Optimization Problems was awarded a Spotlight at ICLR 2025. Congraulations to Vincenzo!
- Grateful to be awarded the 2025 Fellowship in AI Research from the LaCross Institute for Ethical AI in Business. This fellowship will sponsor a project at the intersection of privacy, fairness, and optimization.
- Honored to receive an Outstanding Research Faculty Recongition from the University of Virginia.

- Our work on Learning To Solve Differential Equation Constrained Optimization Problems was accepted to ICLR 2025. Congraulations to Vincenzo!
- Our work on Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion was accepted to NAACL 2025. Congraulations to Jacob and Michael!
- Our work on Differentially Private Data Release on Graphs: Inefficiencies and Unfairness was accepted to AISTATS 2025! With Sen and Juba.
- Our work on Fairness Issues and Mitigations in (Differentially Private) Socio-demographic Data Processes was awarded an oral at AAAI 2025! Congratulations to Joon!

- New preprint: “Multi-Agent Path Finding in Continuous Spaces with Projected Diffusion Models”. See publications for details.
- Several papers accepted at AAAI 2025, main conference and workshops. See publications for details.
- Paper accepted to IEEE SatML 2025! See publications for details.
- Congratulations to Joon for receiving an honorable mention for the 2025 CRA Outstanding Undergraduate Researcher Award!

- We are releasing the first chapter of a new book on Differential Privacy: Differential Privacy Overview and Fundamental Techniques! Check it out if you’d like to learn more about this field.
- I gave a talk on Fairness in constrained ML systems at Ohio State.
- I will give a keynote talk on Differential Privacy and Fairness at the S-HPC workshop at the International Conference of Super Computing (SC)

- New preprint: Learning to Optimize meets Neural-ODE: Real-Time, Stability-Constrained AC OPF! See publications for details.
- The Sixth edition of the Privacy Preserving AI workshop is online! I hope to see many of you there!
- New preprint: End-to-End Optimization and Learning of Fair Court Schedules! See publications for details.
- New preprint: Learning To Solve Differential Equation Constrained Optimization Problems! See publications for details.
- Four papers accepted at NeurIPS-24 workshops!
- I am a member of the chairing committee of the NeurIPS-24 workshop on Algorithmic Fairness through the Lens of Metrics and Evaluation! Hope to seeing you in Vancouver!

- Our paper on constraining the outputs of diffusion models (with guarantees!) has been accepted to NeurIPS 2024! See publications for details.

- New preprint: Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion! See publications for details.
- New preprint: Fairness Issues and Mitigations in (Differentially Private) Socio-demographic Data Processes! See publications for details.
- New preprint: Differentially Private Data Release on Graphs: Inefficiencies and Unfairness! See publications for details.

- Our paper Learning Joint Models of Prediction and Optimization has been accepted to ECAI 2024!
- Our paper Metric Learning to Accelerate Convergence of Operator Splitting Methods for Differentiable Parametric Programming has been accepted to the CDC 2024!

- Invited to attended the National Academy of Science and the Royal Society US-UK Scientific Forum: Scinece in the Age of AI.
- Two papers accepted at ICML workshops!

- I am co-organizing the first Summer School in AI and OR (AISCORE).
- I gave a talk on Disparate impacts of compression in Machine Learning at BuzzRobot.
- Two papers accepted to ICML 2024! On The Fairness Impacts of Hardware Selection in Machine Learning, and Disparate Impact on Group Accuracy of Linearization for Private Inference.

- Our paper End-to-End Learning for Fair Multiobjective Optimization Under Uncertainty has been accepted to UAI 2024!
- I gave a talk on formalizing the principle of data minimization for Machine Learning at the Google Privacy seminar serires.
- Our paper Fairness Increases Adversarial Vulnerability has been accepted to IJCAI 2024!
- Our paper Learning Fair Ranking Policies via Differentiable Optimization of Ordered Weighted Averages has been accepted to FAccT 2024!
- New preprint: Metric Learning to Accelerate Convergence of Operator Splitting Methods for Differentiable Parametric Programming ! See publications for details.

- New preprint: Learning Constrained Optimization with Deep Augmented Lagrangian Methods! See publications for details.
- The fith edition of the Privacy Preserving AI workshop at AAAI-24 has been an exciting event. Check out the website for information about its program!
- Check out my piece on Fairness in The conservation.

- New preprint: End-to-End Learning for Fair Multiobjective Optimization Under Uncertainty! See publications for details.
- New preprint: Projected Generative Diffusion Models for Constraint Satisfaction! See publications for details.
- New preprint: Disparate Impact on Group Accuracy of Linearization for Private Inference! See publications for details.
- New preprint: Learning Fair Ranking Policies via Differentiable Optimization of Ordered Weighted Averages! See publications for details.
- I gave a talk at Amazon Science on the disparate impacts of private machine learning.

- New preprint: Analyzing and Enhancing the Backward-Pass Convergence of Unrolled Optimization! See publications for details.

- Paper accepted to AAAI 2024! See publications for details.
- New preprint: On The Fairness Impacts of Hardware Selection in Machine Learning! See publications for details.
- I am co-organizing the NeurIPS Algorithmic Fairness through the Lens of Times workshop. Hope to see you in NOLA on December 15, 2023!

- New preprint on integrating prediction and optimization via proxy learning! See publications for details.
- I will give talks on the Ethical AI at Toc4Fairness and a keynote at the Comète Workshop on Ethical AI.
- Our paper on disparate impacts arising in energy optimization has been accepted to the NeurIPS 2023 Climate Change AI Workshop!

- I will give talks on the integration of Optimization and Deep Learning at UVA, INFORMS-23, and CCS-23.

- Our paper on data minimization at inference time has been accepted to NeurIPS 2023! See publications for details.
- I am co-organizing the fith edition of the Privacy Preserving AI workshop at AAAI-24. This year edition will have a particular focus on privacy in generative models. Stay tuned!
- Paper accepted in IEEE PES Innovative Smart Grid Technologies. See publications for details.

- I gave a talk on Privacy and Fairness at the IJCAI-23 workshop on Deep Learning Methods for Social Media.
- New survey on integrating prediction and optimization in end-to-end differentiable systems! See publications for details.
- I am co-organizing the Algorithimc Fairness through the Lens of Time workshop at NeurIPS to spark discussions on how a long-term perspective can help build more trustworthy algorithms in the era of generative models.
- New preprint on the disparate impacts arising in energy optimization. See publications for details.
- I have moved to the University of Virginia.

- I gave a talk about the integration of Machine Learning and Optimization at the 2023 ACP Summer School. Youtube link

- Paper accepted to Electric Power Systems Research! See publications for details.
- I gave a talk about Differential Privacy, Foundation and applications in Energy Systems at the DTU PET Summer School. Slides.
- I gave a talk about Machine Learning for Optimization Optimization at the IEEE PES University.

- Two new exciting preprints on privacy and fairness. See publications for details.
- Congratulation to Dr. Cuong Tran on a stellar defense! You can check his work here!
- Our NSF-ENG EPCN proposal Physics Informed Real-time Optimal Power Flow has been funded! We’ll integrate machine learning and physics to optimize power flow, integrate systems dynamics, and increase grid reliability. We’ll work on scalable and robust ML for energy solutions. Thank you, NSF!

- Notable reviewer award ICLR 2023 (link).
- Four papers accepted to IJCAI 2023! See publications for details.

- Our NSF RI CORE proposal End-to-end Learning of Fair and Explainable Schedules for Court Systems. has been funded! We’ll develop differentiable optimization tools for equitable & explainable scheduling and work on changing the pretrial scheduling process to reduce nonappearance and promote fairness in the American Court system! Thank you, NSF!
- I am co-editing a new book on Differential Privacy . It covers topics from foundations, to applications in optimization (including in Census data release, image and video, and medicine), machine learning (including privacy attacks, federated learning, and private algorithms) as well as policy and ethics aspects. Stay tuned!

- I have co-organizing the fourth workshop on Privacy Preserving Artificial Intelligence (PPAI) at AAAI-23.

- Four new exciting preprints on topics including differentiable optimization, data leakage in ML models, differential privacy in language models, and differentially private data disclosure methods. See publications for details.
- Paper accepted to IEEE PES 2023! See publications for details.
- I gave a talk about Differential Privacy and Fairness in Energy Systems at the Grid Science winter school.
- I will be serving as an area chair for FAccT-23 and ECAI-23.
- I will be serving as demo track co-chair for IJCAI-23 and scholarship co-chair for AAMAS-23.
- Paper accepted to AAMAS 2023! See publications for details.