Noam Brown
OpenAI research scientist working on reasoning; creator of Libratus and Pluribus; CICERO co-author
Noam Brown is a Research Scientist at OpenAI, working on reasoning, reinforcement learning, self-play and multi-agent AI, and describes being a foundational contributor to OpenAI's reasoning models such as o1; Brown is also an author of the OpenAI o1 System Card. Before OpenAI, Brown worked at FAIR (Meta) and is a corresponding author of the Science paper introducing CICERO, an agent that plays the negotiation game Diplomacy by combining a language model with strategic reasoning. With a Carnegie Mellon advisor, Brown created the poker AIs Libratus and Pluribus.123
Career
Before graduate school, Brown worked in algorithmic trading from 2006 to 2010 and was a research assistant at the Federal Reserve Board from 2010 to 2012, studying algorithmic trading in financial markets. Brown then earned an MS in robotics (2014) and a PhD in computer science (2020) at Carnegie Mellon University, both advised by Tuomas Sandholm.41
Brown was a research scientist at Facebook AI Research from 2018 to 2023, after a 2017 research internship at DeepMind, and joined OpenAI in 2023. At OpenAI, Brown works on reasoning, reinforcement learning, self-play and multi-agent AI, and is listed among the authors of the o1 and GPT-5 system cards.412
Research
Brown's graduate research with Sandholm focused on imperfect-information games. With Sandholm, Brown built Libratus, described in a 2017 Science paper on beating professional players at heads-up no-limit poker, and Pluribus, the subject of a 2019 Science cover article on multiplayer poker. "Safe and Nested Subgame Solving for Imperfect-Information Games" won a NeurIPS 2017 best paper award.41
At FAIR, Brown's work combined search and learning, including Deep Counterfactual Regret Minimization (ICML 2019) and "Combining Deep Reinforcement Learning and Search for Imperfect-Information Games" (NeurIPS 2020). Work on Diplomacy included no-press Diplomacy agents (2021–2023) and CICERO, which combined a language model with strategic reasoning and was published in Science in 2022.43
Work
OpenAI o1 reasoning models (external site: arxiv.org)Proprietary product
Brown's homepage describes Brown as a foundational contributor to OpenAI's reasoning models (o1); Brown is listed among the authors of the OpenAI o1 System Card (2024).12
CICERO (external site: noambrown.github.io)Published research
Listed as a corresponding author of "Human-level play in the game of Diplomacy by combining language models with strategic reasoning" (Science, 2022), by the Meta Fundamental AI Research Diplomacy Team.31
Libratus and PluribusPublished research
Created the poker AIs Libratus and Pluribus with a Carnegie Mellon advisor; the homepage states Libratus received the IJCAI Marvin Minsky Medal.1
Career timeline
2012
Began graduate research at Carnegie Mellon University with Tuomas Sandholm.4
2017
Published the Libratus paper in Science and won a NeurIPS best paper award for safe and nested subgame solving.4
2018
Joined Facebook AI Research as a research scientist.4
2019
Published "Superhuman AI for Multiplayer Poker", on Pluribus, as a Science cover article.4
2020
Completed a PhD in computer science at Carnegie Mellon University.4
2022
Co-authored the Science paper introducing CICERO for the game Diplomacy.34
2023
Joined OpenAI as a member of technical staff.4
Dec 21, 2024
Listed among the authors of the OpenAI o1 System Card.2
Other documented roles
Roles the sources document but that are not confirmed as current.
- Researcher, Meta Fundamental AI Research Diplomacy Team, FAIR (Meta)13
- Research assistant (2012–2018) and graduate student advised by Tuomas Sandholm (MS 2014, PhD 2020), Carnegie Mellon University4
Honors and recognition
- NeurIPS Best Paper Award · Neural Information Processing Systems Foundation, 20174
- Marvin Minsky Medal for Outstanding Achievements in AI · International Joint Conferences on Artificial Intelligence, 201941
- AAAI/ACM SIGAI Dissertation Award · AAAI and ACM SIGAI, 20204
- Victor Lesser Distinguished Dissertation Award · IFAAMAS, 20204
- Innovators Under 35 · MIT Technology Review, 201941
- ICLR Outstanding Paper Honorable Mention · International Conference on Learning Representations, 20234
Related people
More sourced profiles at the same organization, in the same field, or from the same school.
- Aditya RameshAlso at OpenAI
- Leandro von WerraLanguage models
- Albert GuAlso from Carnegie Mellon
- Alex NicholAlso at OpenAI
- Albert JiangAlso at FAIR
- Alex SmolaAlso from Carnegie Mellon
- Eric MitchellAlso at OpenAI
- Alexander WettigLanguage models
- Alexei A. EfrosAlso from Carnegie Mellon
- Greg BrockmanAlso at OpenAI
- Aravind SrinivasAlso at OpenAI
- Ali FarhadiAlso from Carnegie Mellon
Is this you?
Profiles cover public professional work only, and every line cites a source. If something is out of date, missing or wrong, send a correction with a link to an official page and we will update it. Suggest a correction
Add this badge to your website or GitHub profile README. It links back to this page. It is not an endorsement or a rating.
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/people/noam-brown/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/noam-brown/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Noam Brown — Profile on USASI" height="20"></a>
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.