PeopleofSuperintelligence

Researcher

Oriol Vinyals

Oriol Vinyals is a researcher whose work includes language translation, image captioning, learning from small example sets, and game-playing agents. His university training combined mathematics and telecommunications at UPC, followed by study at UC San Diego and a doctorate at Berkeley. He co-authored sequence-to-sequence learning and worked with the teams behind StarCraft II research tools and AlphaStar. The 2019 AlphaStar announcement described a specific demonstration with interface conditions that differed from ordinary human play. His later contributions include Gemini. In November 2025, UPC gave him an honorary doctorate and described him as a DeepMind vice president.

Works
218
Citations
163,034
h-index
90

Publication numbers from OpenAlex, as of 14 October 2026.

Photo of Oriol Vinyals

Timeline

  1. September 2014

    Turning one sentence into another

    Vinyals co-authored sequence-to-sequence learning in 2014.

  2. November 2014

    Giving an image a caption

    Vinyals co-authored neural image captioning in November 2014.

  3. June 2016

    Learning from a small example set

    Vinyals co-authored Matching Networks in 2016.

  4. August 2017

    Opening a strategy game to research

    Vinyals helped introduce StarCraft II research tools in 2017.

  5. 2019

    The AlphaStar demonstration

    Vinyals helped present the AlphaStar demonstration in 2019.

  6. 2023–2025

    Models that combine different inputs

    Vinyals co-authored Gemini in 2023; UPC listed his vice president role in 2025.

Sources and credits

Photo: UPC CFIS profile. License: Not stated by source. Source: cfis.upc.edu. Cropped to a square and resized. Ask us to remove or replace this photo.

  1. UPC: Oriol Vinyals honorary doctorate upc.edu. Checked 11 October 2026.
  2. Sequence to Sequence Learning with Neural Networks arxiv.org. Checked 11 October 2026.
  3. Show and Tell: A Neural Image Caption Generator arxiv.org. Checked 11 October 2026.
  4. Distilling the Knowledge in a Neural Network arxiv.org. Checked 11 October 2026.
  5. Matching Networks for One Shot Learning arxiv.org. Checked 11 October 2026.
  6. DeepMind and Blizzard open StarCraft II as an AI research environment deepmind.google. Checked 11 October 2026.
  7. AlphaStar: Mastering the real-time strategy game StarCraft II deepmind.google. Checked 11 October 2026.
  8. Gemini: A Family of Highly Capable Multimodal Models arxiv.org. Checked 11 October 2026.