Model-Agnostic Signal Discovery with Machine Learning: Bridging the Gap Between Theory and Practice

2026-05-29

AI Summary

Traditional approaches to new physics searches rely heavily on specific theoretical models, limiting their ability to explore unknown signal spaces—particularly in the absence of strong prior guidance. This work establishes a unified conceptual framework that systematically integrates mainstream model-agnostic anomaly detection and signal discovery methods, especially those driven by artificial intelligence in high-energy physics and related fields. Emphasizing broad exploration of complex scientific data over optimization for predefined hypotheses, the study proposes rigorous validation and interpretability strategies to bridge the gap between theoretical formulation and practical deployment. By providing researchers with a principled reference, this framework advances the standardized application and reliable interpretation of model-agnostic techniques, thereby substantially enhancing the discovery potential of modern scientific experiments.

Abstract

Searches for new phenomena in complex scientific data are predominantly model-dependent, optimized for specific hypotheses, and therefore limited in their coverage of the space of possible signals. Recently, new AI-based model-agnostic search strategies, many of which have been pioneered in high-energy physics, have been proposed which provide a complementary paradigm, prioritizing broad exploration over tailored analyses. These techniques offer an opportunity to enhance the overall discovery potential of modern experiments, especially in regimes where theoretical guidance is scarce. In this document, we review the conceptual framework behind the main classes of AI-based model-agnostic strategies. We discuss the potential pitfalls of these methods, and strategies for their validation and interpretation. We aim for this document to serve as a useful reference both for practitioners and for researchers interested in learning more about these model-agnostic search strategies.

Problem

• model-agnostic
• signal discovery
• machine learning
• new physics
• anomaly detection

Innovation

• model-agnostic
• signal discovery
• machine learning
• anomaly detection
• high-energy physics