Welcome to my personal page! I am a 5th year PhD student at MIT working in LIDS with Tamara Broderick. I am broadly interested in AI4Science, in particular generative models and uncertainty quantification for biological and environmental applications. More precisely, I develop probabilistic machine learning methods for spatiotemporal data that embed physical and biological structure into their design. During my PhD, I closely collaborated with domain scientists, including oceanographers, biologists, and atmospheric chemists.
This past summer I interned in the Center for Computational Mathematics at the Flatiron Institute in New York City where I worked on Simulation-based Inference with Diana Cai. I previously interned as a Machine Learning researcher in the Health AI team at Apple. Before joining MIT, I completed a MSc in Data Science at Bocconi University advised by Igor Prünster. Prior to that, I obtained a BSc in Economics and Computer Science advised by Massimo Marinacci.
Besides research, I have always been very passionate about having an active role in the communities I am part of. I have previously served as co-President of the MIT EECS Graduate Students’ Association. I am currently serving as Program Chair and organizer of the Bayesian Decision Making and Uncertainty workshop at NeurIPS and organizer of the Virtual Seminar Series on Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems.
Feel free to contact me via email at renb at mit dot edu.
I am on the 2025–2026 postdoc job market — please reach out if you want to chat!
- Generative Models
- Bayesian Inference
- Uncertainty Quantification
- Trajectory and SDE inference in Computational Biology
- Spatiotemporal Machine Learning for Environmental Applications
-
PhD in Electrical Engineering & Computer Science, 2026 (expected)
Massachusetts Institute of Technology
-
MSc in Data Science & Business Analytics, 2021
Bocconi University
-
BSc in Economics & Computer Science, 2019
Bocconi University
Recent news
-
September 2025: Our paper Smooth Sailing: Lipschitz-Driven Uncertainty Quantification for Spatial Association was accepted to NeurIPS 2025! See you in San Diego 🏄🏻♂️
-
August 2025: I am very grateful to have received the Panasonic Well AI for Health graduate fellowship to support me for my final year of PhD.
-
April 2025: Super excited to join the Center for Computational Mathematics at the Flatiron Institute for this coming summer!
-
February 2025: Our new work on quantifying uncertainty in spatial linear models under misspecification and covariate shift was just posted on arXiv!
-
January 2025: Our paper Multi-marginal Schrödinger Bridges with Iterative Reference Refinement was accepted as an oral at AISTATS 2025! See you in Thailand 🇹🇭
-
September 2024: We just posted on arXiv a new exciting work on evaluating PM2.5 forecasts from the perspective of individual decision making.
-
August 2024: Our new work on reconstructing particle trajectories with Schrödinger Bridges was just posted on arXiv – check it out!
-
July 2024: I am organizing a NeurIPS workshop on Bayesian Decision Making and Uncertainty – please consider submitting your research to it :)
-
February 2024: I won the Outstanding Student Presentation Award at the AGU 2023 Annual Meeting
-
June 2023: I have been selected as a Student Poster Award winner at the 36th New Englang Statistics Symposium.
-
May 2023: I am excited to share I will be in Seattle for the summer interning at Apple, doing research in the Health AI team!
-
February 2023: I won the Best Presentation Award for the talk on ‘‘Gaussian Processes at the Helm(holtz)’’ at 28th Annual LIDS Student Conference, Optimization and Algorithms Session :)
-
July 2022: I have been selected as a Poster Award winner in the category BayesComp/j-ISBA at ISBA 2022.
-
June 2022: I won the Microsoft Award for the best talk at BAYSM 2022!
Publications
Preprints, * denotes equal contribution
-
Berlinghieri, R.*; Shen, Y.*; Broderick, T. Oh SnapMMD! Forecasting stochastic dynamics beyond the Schrödinger bridge’s end. Under review.
-
Berlinghieri, R.*; Burt, D. R.*, Giani, P.; Fiore, A.; Broderick, T. Are Hourly PM2.5 Forecasts Sufficiently Accurate to Plan Your Day? Individual Decision Making in the Face of Increasing Wildfire Smoke. Under review.
-
Burt, D. R.; Berlinghieri, R.; Broderick, T. Wrong Model, Right Uncertainty: Spatial Associations for Discrete Data with Misspecification. In preparation.
Publications
-
Burt, D. R.*; Berlinghieri, R.*; Bates, S.; Broderick, T. Smooth Sailing: Lipschitz-Driven Uncertainty Quantification for Spatial Association. To appear in The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025).
-
Shen, Y.*; Berlinghieri, R.*; Broderick, T. Multi-marginal Schrödinger Bridges with Iterative Reference Refinement . Oral presentation at 28th International Conference on Artificial Intelligence and Statistics 2025 (AISTATS 2025)
-
Berlinghieri, R.; Trippe, B. L.; Burt, D. R.; Giordano, R.; Srinivasan, K.; Özgökmen, T.; Junfei, X.; & Broderick, T. Gaussian processes at the Helm (holtz): A more fluid model for ocean currents. In International Conference on Machine Learning (ICML 2023).
-
Berlinghieri, R.; Krajbich, I.; Maccheroni, F.; Marinacci, M.; Pirazzini, M. Measuring utility with diffusion models. Science Advances 9 (34), 2023.
-
Jing, B.*; Corso, G.*; Berlinghieri, R.; & Jaakkola, T. (2022). Subspace diffusion generative models. In European Conference on Computer Vision (ECCV 2022).
Workshop Publications
-
Burt, D. R.; Berlinghieri, R.; Broderick, T. Wrong Model, Right Uncertainty: Spatial Associations for Discrete Data with Misspecification. To appear in NeurIPS 2025 AI4Science Workshop. [Spotlight]
-
Berlinghieri, R.*; Shen, Y.*; Broderick, T. Beyond Schrödinger Bridges: A Least-Squares Approach for Learning Stochastic Dynamics with Unknown Volatility. In Frontiers in Probabilistic Inference workshop at ICLR 2025.
-
Shen, Y.*; Berlinghieri, R.*; Broderick, T. Learning a vector field from snapshots of unidentified particles rather than particle trajectories.In AI4Differential Equations In Science workshop at ICLR 2024.
-
Berlinghieri, R.; Trippe, B. L.; Burt, D. R.; Giordano, R.; Srinivasan, K.; Özgökmen, T.; Junfei, X.; Broderick, T. Gaussian processes at the Helm(holtz): A better way to model ocean currents. In NeurIPS 2022 Workshop on Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems. [Contributed Talk]
Contact
- renb [at] mit [dot] edu
- 32-D740 - 32 Vassar Street, Cambridge, MA 02139
- Contact me