Houssam Zenati
Research Fellow @ Gatsby Computational Neuroscience Unit, UCL
- London, UK
- Google Scholar
- GitHub
I develop machine learning algorithms and statistical theory for learning from feedback, understanding intervention effects, and designing adaptive experiments. My research progresses from learning which interventions to choose, to establishing what they change, to deciding what to try next—when good decisions and informative experiments need not favor the same assignments.
Research overview All publications
About Me
I am a Research Fellow at the Gatsby Computational Neuroscience Unit at University College London, where I work with Arthur Gretton.
Previously, I was a postdoctoral researcher in the MIND team at Inria, working on causal inference and mediation analysis. I completed my PhD jointly with Inria Thoth and the Criteo AI Lab on offline policy learning and sequential learning. Earlier, at the Institute for Infocomm Research in Singapore, I worked on deep generative models and anomaly detection.
Feel free to reach out if you wish to collaborate, exchange ideas, or seek Master's thesis supervision. Contact: (first initial).(last name) [at] ucl.ac.uk.
Selected Publications
Learning to Intervene
- Sequential Counterfactual Risk Minimization
Houssam Zenati, Eustache Diemert, Matthieu Martin, Julien Mairal, Pierre Gaillard.
ICML, 2023.
[Paper] · [Code]
Inference on Intervention Effects
- Semiparametric Efficient Test for Interpretable Distributional Treatment Effects
Houssam Zenati, Arthur Gretton.
NeurIPS, 2026.
[Paper]
Adaptive Experimentation
- Efficient Inference after Directionally Stable Adaptive Experiments
Z. Shen*, H. Zenati*, N. Kallus, A. Gretton, K. Khamaru, A. Bibaut.
Preprint, 2026.
[arXiv]
* Equal contribution.
News
- September 2026: Our paper Semiparametric Efficient Test for Interpretable Distributional Treatment Effects was accepted at NeurIPS 2026.
- June 2026: Lead organiser of the Advances in Adaptive Experimentation (AAE) Workshop at the Gatsby Unit, UCL.
- June 2026: Our paper Fast Best-in-Class Regret for Contextual Bandits was accepted at UAI 2026.