About me

My name is Jean Pouget-Abadie, but non-French speakers can call me John. I am a research scientist at Google DeepMind in New York on the Gemini Large Scale Pretraining team. Previously, I led a small research team within the Algorithms & Optimization group within Google Research. Before joining Google, I completed a PhD in Computer Science at Harvard University, advised by Edoardo Airoldi and Salil Vadhan, where I was a 2017-2018 Siebel scholar. Prior to that, I was an undergraduate at Ecole Polytechnique, Paris. At the time, I looked into using neural networks to generate new distributions from data.

Resume Google Scholar DBLP

Mini-CV

  • Senior Staff Research Scientist at Google DeepMind
    with Paul Michel on the Gemini Large Scale Pretraining team (2026-)

  • Senior Staff Research Scientist at Google
    with Vahab Mirrokni in the algorithms research group (2018-2026)

  • Spotify summer internship
    with the Discover Weekly team (music recommendation) (2017)

  • Facebook summer internship
    with Udi Weinsberg on the Core Data Science Team (2015)

  • Harvard University
    PhD program in Computer Science (2014-2018)

  • MILA, Université de Montréal
    with Yoshua Bengio in Deep Learning (2014)

  • École Polytechnique
    Diplôme d'ingénieur (2011-2014)

Selected Publications

  • GenAI

    • Generative Adversarial Networks. Ian Goodfellow, Jean Pouget-Abadie, Medhi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. [arXiv, NeurIPS'14]

    • Segmentation for Neural Machine Translation. Jean Pouget-Abadie, Dzmitry Bahdanau, Bart van Merriënboer, Kyung Hyun Cho, and Yoshua Bengio. SSST-8 @ EMNLP 2014 [arXiv]

  • Bipartite Experiments

    • Design and analysis of bipartite experiments under a linear exposure-response model. Christopher Harshaw, Fredrik Savje, David Eisenstat, Vahab Mirrokni, Jean Pouget-Abadie. [arXiv, EJS'23]

    • Cluster Randomized Designs for One-Sided Bipartite Experiments. Jennifer Brennan, Vahab Mirrokni, Jean Pouget-Abadie. [arXiv, NeurIPS'22]

    • Variance reduction of bipartite experiments through Correlation Clustering. Jean Pouget-Abadie, Kevin Aydin, Warren Schudy, Kay Brodersen, Vahab Mirrokni. [NeurIPS'19]

  • Small-samples and Geographical Experiments

    • Causal Bootstrap for General Randomized Designs. Jennifer Brennan, Sébastien Lahaie, Adel Javanmard, Nick Doudchenko, Jean Pouget-Abadie [arXiv]

    • Randomized Experimental Design via Geographic Clustering. David Rolnick, Kevin Aydin, Jean Pouget-Abadie, Shahab Kamali, Vahab Mirrokni, Amir Najmi. [KDD'19]

    • Synthetic Design: An Optimization Approach to Experimental Design with Synthetic Controls. Nick Doudchenko, Khashayar Khosravi, Jean Pouget-Abadie, Sébastien Lahaie, Miles Lubin, Vahab Mirrokni, Jann Spiess, Guido Imbens. [NeurIPS'21]

    • Randomized Experimental Design via Geographic Clustering. David Rolnick, Kevin Aydin, Jean Pouget-Abadie, Shahab Kamali, Vahab Mirrokni, Amir Najmi. [KDD'19]

  • Dealing with Interference

    • Modeling Interference Using Experiment Roll-Out. Ariel Boyarsky, Hongseok Namkoong, Jean Pouget-Abadie. [arXiv, EC'23]

    • Testing for Arbitrary Interference on Experimentation Platforms. Jean Pouget-Abadie, Guillaume Saint-Jacques, Martin Saveski, Weitao Duan, Ya Xu, Souvik Ghosh, and Edoardo M. Airoldi. [arXiv, Biometrika'19]

    • Dealing with Interference on Experimentation Platforms. Jean Pouget-Abadie. Harvard 2018. [Dissertation]

    • Detecting Network Effects: Randomizing over Randomized Experiments. Martin Saveski, Jean Pouget-Abadie, Guillaume Saint-Jacques, Weitao Duan, Ya Xu, Souvik Ghosh, and Edoardo M. Airoldi. [KDD'17]

Mentorship

I've had the privilege of mentoring some incredible Googlers, interns, and AI residents over the years: Khashayar Khosravi, Nick Doudchenko, Chris Harshaw, Jennifer Brennan, Evan Munro, Ian Waudby-Smith, Roshni Sahoo, Lalit Jain, Clayton Sanford, Rudrajit Das, Negar Foroutan, and Yuchen Li

Misc.

  • [2023] CODE@MIT Plenary on Bipartite Experiments here.

  • [2017] Introduction to non-independent A/B tests here.

  • [2015] Submodular maximization reading notes here.