PeopleofSuperintelligence

Researcher

Michael Albergo

Michael Albergo is an applied mathematician whose work connects generative AI with scientific sampling. In 2019, he coauthored a flow-based sampler for lattice field theory; a 2025 Journal of Machine Learning Research paper he coauthored unified flow and diffusion models. He completed a Ph.D. at New York University in 2024 and received an AI2050 Early Career Fellowship in 2025 for research on controllable generative models. In March 2026, he presented research on reward alignment for flow and diffusion models. Since July 2026, he is an assistant professor of applied mathematics at Harvard and a Kempner Institute Investigator.

Works
52
Citations
1,130
h-index
15

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2019

    Flow-based Monte Carlo sampling

    In 2019, he coauthors “Flow-based generative models for Markov chain Monte Carlo in lattice field theory,” which proposes a machine-learned flow-based update scheme for Monte Carlo sampling in lattice field theories.

  2. 2022

    Continuous-time normalizing flows

    In 2022, he coauthors “Building Normalizing Flows with Stochastic Interpolants,” proposing a continuous-time flow between any pair of base and target probability densities.

  3. 2025

    Unifying flows and diffusions

    In 2025, he publishes “Stochastic Interpolants: A Unifying Framework for Flows and Diffusions” in the Journal of Machine Learning Research. The paper introduces a class of generative models that unifies flow-based and diffusion-based methods.

  4. 2024

    A PhD and Harvard appointment

    In 2024, he completes a Ph.D. at New York University. Harvard announces that he will join as a Kempner Institute Investigator and a Harvard SEAS assistant professor of applied mathematics in July 2026.

  5. November 2025

    An AI2050 Early Career Fellowship

    In November 2025, Schmidt Sciences names him an AI2050 Early Career Fellow. His project studies flow and diffusion models and seeks ways to make them more efficient and controllable for generative control of dynamical, living matter.

  6. March 2026

    Reward alignment for generative models

    In March 2026, he presents at the Flatiron Institute on scalable reward alignment for flow and diffusion models, including fine-tuning during training and steering during inference.

  7. Since July 2026

    Assistant professor at Harvard

    Since July 2026, he is an assistant professor of applied mathematics at Harvard and a Kempner Institute Investigator.

Career

  1. Since 2026Harvard UniversityAssistant Professor of Applied Mathematics
  2. Since 2026Harvard UniversityInvestigator

Education

  1. 2024New York UniversityPhD

From the biography.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Michael Albergo yet. Send us a free-licensed photo.

  1. The 37th International Symposium on Lattice Field Theory (Lattice 2019): Flow-based generative models for MCMC in lattice field theory indico.cern.ch. Checked 27 October 2026.
  2. Building Normalizing Flows with Stochastic Interpolants arxiv.org. Checked 27 October 2026.
  3. Stochastic Interpolants: A Unifying Framework for Flows and Diffusions jmlr.org. Checked 27 October 2026.
  4. Michael Albergo - AI2050 ai2050.schmidtsciences.org. Checked 27 October 2026.
  5. Kianté Brantley, Yilun Du and Michael Albergo to join the Kempner Institute news.harvard.edu. Checked 27 October 2026.
  6. Schmidt Sciences Awards Early-Career Fellowships to Michael Albergo, Melanie Weber seas.harvard.edu. Checked 27 October 2026.
  7. Machine Learning at the Flatiron Institute Seminar: Michael Albergo simonsfoundation.org. Checked 27 October 2026.
  8. Michael Albergo - Kempner Institute kempnerinstitute.harvard.edu. Checked 27 October 2026.