Advances in Adaptive Experimentation is a 2-day workshop hosted by the Gatsby Computational Neuroscience Unit at University College London on June 18–19, 2026.

The workshop brings together researchers working at the interface of adaptive experimentation, causal inference, online learning, bandits, and modern statistical methods for sequential decision-making. It is intended as a focused technical meeting point for theory, methodology, and discussion across communities concerned with learning, inference, and decision-making from adaptively collected data.

The programme will feature invited talks, informal discussion, an open problems session, and a poster session. Topics include exploration-exploitation trade-offs, stability conditions for adaptive data collection, regret minimization, best-arm identification, structured feedback, identification, nuisance-robust methods, and semiparametric efficiency theory.

The open problems session will provide space for concise technical questions, conjectures, limitations, and unresolved obstacles that may benefit from collective discussion. The poster session will highlight recent, ongoing, or discussion-stage theoretical work in adaptive experimentation and related foundations of sequential decision-making.

Please also see the "More Info Tab"