PeopleofSuperintelligence

Researcher

Daniel Ramage

Also known as D. Ramage

Daniel Ramage is a Google researcher whose work helped bring federated learning to mobile devices. He studied computer science and mathematics at MIT, completed a Stanford doctorate in natural language processing, and helped develop topic models for labeled text. At Google, he worked with collaborators on federated language models for Gboard and privacy tools for decentralized data. As of October 2026, he works at Google on federated learning and privacy technologies.

Now at
Google
Works
66
Citations
90,601
h-index
32
Rankings
#94 by citations and #60 by citations since 2021
Badges

Publication numbers from OpenAlex, as of 19 October 2026.

Timeline

  1. By 2011

    A Stanford NLP doctorate

    He completes a computer-science PhD at Stanford in 2011 under Christopher D. Manning in the Stanford Natural Language Processing Group. Before Stanford, he studies computer science and math at MIT, works on data mining at IBM's Zurich Research Lab, and works with Susan Dumais at Microsoft Research in fall 2009.

  2. 2011

    Partially labeled topic models

    In 2011, he coauthors Partially Labeled Topic Models for Interpretable Text Mining, which introduces two partially supervised models, PLDA and PLDP, that discover hidden topics within human-labeled text and improve interpretability.

  3. Since 2011

    Google research scientist

    By December 2011, he is a research scientist at Google in Mountain View. His work focuses on federated learning and privacy technologies.

  4. 2016–2017

    Introducing federated learning

    In 2016, he and four collaborators introduce federated learning through a method that trains a shared model from locally computed updates on mobile devices. In 2017, he and Brendan McMahan describe how Gboard uses the approach for query suggestions while training data stays on-device.

  5. 2018

    Mobile keyboard prediction

    In 2018, he coauthors a study that trains a recurrent neural language model on mobile devices for next-word prediction in a virtual keyboard; the federated model achieves higher prediction recall and keeps sensitive training examples on-device.

  6. 2020–2021

    Privacy for federated data

    In 2020, he and Stefano Mazzocchi describe federated analytics as a way to study decentralized data, including counts and rates. In 2021, he coauthors the ACM Queue article Federated Learning and Privacy, which outlines privacy-preserving machine learning and data science on decentralized data.

  7. March 2025

    Confidential federated analytics

    In March 2025, he and Timon Van Overveldt introduce confidential federated analytics, a method that makes server-side processing inspectable and limits access to approved analyses. Ramage is a Research Director at Google.

  8. As of October 2026

    Federated learning at Google

    As of October 2026, he works at Google on federated learning and privacy technologies. At the Simons Institute, he discusses open problems in provably private federated learning.

Career

  1. Since 2011GoogleResearch Scientist
  2. 2009Microsoft Research

Education

  1. 2011Stanford UniversityPhD
  2. Massachusetts Institute of Technology

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Cytoscape: A Software Environment for Integrated Models of Biomolecular Interaction Networks2003 · Genome Research · 51,679 citations
  2. Communication-Efficient Learning of Deep Networks from Decentralized Data2016 · 14,215 citationsWith Blaise Agüera y Arcas
  3. Advances and Open Problems in Federated Learning2020 · Foundations and Trends® in Machine Learning · 7,624 citationsWith Dawn Song, Farinaz Koushanfar, Mehryar Mohri, Qiang Yang, Ramesh Raskar and Sanmi Koyejo
  4. Practical Secure Aggregation for Privacy-Preserving Machine Learning2017 · ACM SIGSAC Conference on Computer and Communications Security (CCS) · 3,908 citations
  5. Towards Federated Learning at Scale: System Design2019 · 2,283 citations

From OpenAlex, as of 19 October 2026.

  1. Google has the highest combined h-index among AI organizationsIncludes Google. Our estimates of combined and median h-index at AI organizations, using current affiliations and OpenAlex records.
  2. Google alumni founded or led 87 other organizations, the most among the labs we compareIncludes Google. We trace 212 organizations founded or led by AI lab alumni, and 223 career moves between the labs.
  3. Google had the most people shaping AI every year from 2014 to 2025Includes Google. Our estimates of which organizations employed the most people shaping AI from 1961 to 2025, with a year-by-year chart.
  4. Of 462 people shaping AI who worked at Google, 190 work there nowIncludes Google. Our estimates of where former Google and Google DeepMind people work now, including moves within Google.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

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

  1. Daniel Ramage nlp.stanford.edu. Checked 19 October 2026.
  2. Partially Labeled Topic Models for Interpretable Text Mining microsoft.com. Checked 19 October 2026.
  3. Daniel Ramage research.google. Checked 19 October 2026.
  4. Communication-Efficient Learning of Deep Networks from Decentralized Data arxiv.org. Checked 19 October 2026.
  5. Federated Learning: Collaborative Machine Learning without Centralized Training Data research.google. Checked 19 October 2026.
  6. Federated Learning for Mobile Keyboard Prediction arxiv.org. Checked 19 October 2026.
  7. Federated Analytics: Collaborative Data Science without Data Collection research.google. Checked 19 October 2026.
  8. Discovering new words with confidential federated analytics research.google. Checked 19 October 2026.
  9. Toward Provably Private Federated Learning simons.berkeley.edu. Checked 19 October 2026.