PeopleofSuperintelligence

Researcher

Finale Doshi-Velez

Finale Doshi-Velez is a computer scientist whose work connects interpretable machine learning and reinforcement learning with decisions in health care. Her education includes an MSc at Cambridge as a Marshall Scholar and a PhD at MIT. Her 2017 papers addressed model explanations and rigorous interpretability evaluation, and she was named a Sloan Research Fellow in 2018. She coauthored a 2019 roadmap for responsible machine learning in health care and a 2022 framework for interpretable reinforcement learning in ICU hypotension management; in 2024, she argued that decision-support AI should give physicians several treatment options with expected pros and cons. As of October 2026, she is the Herchel Smith Professor of Computer Science at Harvard SEAS and heads the Data to Actionable Knowledge group.

Works
300
Citations
13,381
h-index
47

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2011

    MIT classes on uncertainty

    At MIT, she studies statistical modeling and robust decision-making as a computer science graduate student. In 2011, she teaches MIT ESP classes on uncertainty and Markov chain Monte Carlo.

  2. 2013

    Harvard postdoc and AI recognition

    In 2013, she is a postdoctoral research associate at Harvard SEAS and Harvard Medical School, based primarily at the Center for Biomedical Informatics. Her projects apply probabilistic modeling to autism trajectories and diabetes-treatment records, and IEEE Intelligent Systems names her among its “AI’s 10 to Watch” researchers.

  3. 2017

    Right for the Right Reasons

    With Andrew Slavin Ross and Michael C. Hughes, she coauthors the 2017 IJCAI paper “Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations.” It uses the gradient of the class label with respect to the input to explain model predictions.

  4. 2017

    Defining interpretable machine learning

    With Been Kim, she coauthors a 2017 position paper that defines interpretability and proposes a taxonomy for rigorous evaluation.

  5. 2018

    A Sloan Research Fellowship

    She is named an Alfred P. Sloan Research Fellow in 2018 while she is an assistant professor of computer science at SEAS. Her research applies probabilistic machine learning to health care and law.

  6. 2019

    Responsible ML for health care

    She coauthors a 2019 Nature Medicine roadmap for responsible machine learning in health care. It presents guidelines to accelerate translation of machine-learning interventions, with engaged stakeholders and a process from problem formulation to deployment.

  7. 2021

    Full professor with tenure

    Harvard approves her as a full professor with tenure at SEAS in 2021. Her work focuses on probabilistic machine learning for health care, tools to support human decisions, and interpretable models.

  8. 2022

    Interpretable RL for ICU hypotension

    In 2022, she coauthors an interpretable reinforcement learning framework for ICU hypotension management. It identifies clinical contexts where clinicians choose different treatments, traces their effects, and develops recommendations that experts can inspect before deployment.

  9. 2024

    AI options for clinical decisions

    In 2024, she argues that AI decision support should offer physicians several treatment options with expected pros and cons to invite them to engage with the system.

  10. As of October 2026

    Professor and DtAK leader

    She is the Herchel Smith Professor of Computer Science at Harvard SEAS and heads the Data to Actionable Knowledge group. In 2026, she coauthors a study on consumer wearable APIs for predicting substance use and craving.

Career

  1. NowHarvard UniversityHerchel Smith Professor of Computer Science
  2. NowHarvard UniversityHead of the Data to Actionable Knowledge group
  3. From 2021Harvard UniversityFull Professor with Tenure
  4. Harvard UniversityPostdoctoral Research Associate
  5. Harvard UniversityAssistant Professor of Computer Science

Education

  1. University of CambridgeMSc
  2. Massachusetts Institute of TechnologyPhD

From the biography.

Checked career sources: kempnerinstitute.harvard.edu.

Colleagues

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

Sources and credits

Photo: We have no published photo of Finale Doshi-Velez yet. Send us a free-licensed photo.

  1. MIT ESP - Finale Doshi-Velez esp.mit.edu. Checked 27 October 2026.
  2. Finale Doshi-Velez named among “AI’s 10 to Watch” seas.harvard.edu. Checked 27 October 2026.
  3. Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations | IJCAI ijcai.org. Checked 27 October 2026.
  4. [1702.08608] Towards A Rigorous Science of Interpretable Machine Learning arxiv.org. Checked 27 October 2026.
  5. Finale Doshi-Velez and Yaron Singer named Sloan Research Fellows seas.harvard.edu. Checked 27 October 2026.
  6. Do no harm: a roadmap for responsible machine learning for health care | Nature Medicine nature.com. Checked 27 October 2026.
  7. Finale Doshi-Velez granted tenure seas.harvard.edu. Checked 27 October 2026.
  8. An interpretable RL framework for pre-deployment modeling in ICU hypotension management pmc.ncbi.nlm.nih.gov. Checked 27 October 2026.
  9. Fifteen Questions: Finale Doshi-Velez on AI Decision Making, Novel Writing and Unicorns | The Harvard Crimson thecrimson.com. Checked 27 October 2026.
  10. Alissia Di Maria, S.M. '26: Diving into the data of wearable tech seas.harvard.edu. Checked 27 October 2026.
  11. People - Center for Human-driven AI Research and Methods (CHARM) at Harvard charm.seas.harvard.edu. Checked 27 October 2026.
  12. Research - Center for Human-driven AI Research and Methods (CHARM) at Harvard charm.seas.harvard.edu. Checked 27 October 2026.
  13. Finale Doshi-Velez - Kempner Institute kempnerinstitute.harvard.edu. Checked 27 October 2026.