QCD-aware recursive neural networks for jet physics

By Gilles Louppe, Kyunghyun Cho, C. Becot, Kyle Cranmer. 2017

201 citations recorded by the source corpus.

Recent progress in applying machine learning for jet physics has been built upon an analogy between calorimeters and images. In this work, we present a novel class of recursive neural networks built instead upon an analogy between QCD and natural languages. In the analogy, four-momenta are like words and the clustering history of sequential recombination jet algorithms is like the parsing of a sentence. Our approach works directly with the four-momenta of a variable-length set of particles, and the jet-based tree structure varies on an event-by-event basis. Our experiments highlight the flexibility of our method for building task-specific jet embeddings and show that recursive architectures are significantly more accurate and data efficient than previous image-based networks. We extend the analogy from individual jets (sentences) to full events (paragraphs), and show for the first time an event-level classifier operating on all the stable particles produced in an LHC event.

Subfields: Artificial Intelligence, Nuclear and High Energy Physics, Model Architectures, AI for Physical Sciences

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. E. Metodiev
2. Jesse Thaler
3. Patrick T. Komiske
4. Benjamin Nachman
5. A. Larkoski
6. I. Moult
7. Jonathan Shlomi
8. Jean-Roch Vlimant

In Nuclear and High Energy Physics

1. Benjamin Nachman
2. I. Moult
3. E. Metodiev
4. Patrick T. Komiske
5. A. Larkoski
6. Jesse Thaler
7. Kyle Cranmer
8. G. Soyez

In Model Architectures

1. E. Metodiev
2. Jesse Thaler
3. Jonathan Shlomi
4. Kyle Cranmer
5. I. Cirac
6. Leslie Vogt-Maranto
7. Laurent Daudet
8. Naftali Tishby

In AI for Physical Sciences

1. A. Larkoski
2. E. Metodiev
3. Patrick T. Komiske
4. Benjamin Nachman
5. Kyle Cranmer
6. I. Moult
7. Jesse Thaler
8. Peter W. Battaglia

Citations in the corpus

• Machine learning and the physical sciences (2019, methodology): “For example, motivated by techniques in natural language processing, recursive networks were designed that operate over tree-structures created from a class of jet clustering algorithms (Louppe et al., 2017a).”
• Jet substructure at the Large Hadron Collider: A review of recent advances in theory and machine learning (2017, background): “The first application of recurrent NNs to jet physics was in the context of boosted boson tagging [542] and has also been applied to quark-versus-gluon tagging [543].”
• Energy flow networks: deep sets for particle jets (2018, methodology): “We did not consider alternative particle orderings for the RNN, though other investigations have found that performance is robust to choices in particle ordering [15, 42].”
• Graph neural networks in particle physics (2020, background): “[24] Gilles Louppe, Kyunghyun Cho, Cyril Becot, and Kyle Cranmer.”