28 September 2026 to 2 October 2026
University of Oxford
Europe/London timezone

Scaling Neural Simulation Based Inference to Large Datasets

Not scheduled
20m
University of Oxford

University of Oxford

Standard slot New techniques, tools, and generators New techniques, tools and generators

Speaker

Chris Pollard (University of Warwick (GB))

Description

Neural simulation-based inference (NSBI) in HEP predominantly relies on event-level likelihood or likelihood-ratio estimators. In non-trivial settings, these require expensive post-hoc profiling or marginalization over nuisance parameters to perform dataset-wide inference. We introduce PAIRS, a two-stage procedure that learns an event representation from small datasets ($N \le 2$). Under mild regularity assumptions, this results in an observable that is information-preserving and has a simple, fast composition rule, with the potential to resolve a fundamental scaling bottleneck for NSBI at the LHC. Validated on several benchmarks, including a resonance search inspired by LHC bump hunts, PAIRS empirically matches or outperforms dataset-wide training at a fraction of the compute for a variety of problems and offers a practical path to enable amortized inference over large datasets.

Author

Chris Pollard (University of Warwick (GB))

Presentation materials

There are no materials yet.