Name: Machine Learning approaches in Lattice QCD - An interdisciplinary exchange
Start: 2023-02-27T08:00:00+01:00
End: 2023-03-03T20:00:00+01:00
Location: Institute for Advanced Study of the Technische Universität München

Machine Learning approaches in Lattice QCD - An interdisciplinary exchange

from Monday, 27 February 2023 (08:00) to Friday, 3 March 2023 (20:00)

Monday, 27 February 2023

08:30 Registration
Registration
08:30 - 09:10
09:10 Welcome and aims of the workshop - Nora Brambilla
Welcome and aims of the workshop
- Nora Brambilla
09:10 - 09:25
09:25 Machine learning is ubiquitous - Maria-Paola Lombardo
Machine learning is ubiquitous
- Maria-Paola Lombardo
09:25 - 09:55
09:55 Generative flow methods
09:55 - 10:45
Contributions

09:55 Aspects of scaling and scalability for flow-based samplers - Daniel Hackett

10:20 Learning trivializing flows - David Albandea
10:45 Coffee break
Coffee break
10:45 - 11:15
11:15 Generative flow methods
11:15 - 12:40
Contributions

11:15 Fourier-Flow model generating Feynman paths - Lingxiao Wang

11:40 Stochastic normalizing flows for lattice field theory - Elia Cellini

12:00 Panel discussion - Chair: Sinead Ryan Members: Simone Bacchio Daniel Hackett Maria Paola Lombardo Alberto Ramos
13:00 Lunch
Lunch
13:00 - 14:30
14:30 ML for particle physics
14:30 - 16:00
Contributions

14:30 Discovering mathematical structures with neural networks - Sven Krippendorf

14:50 Machine Learning at IceCube - Rasmus Orsoe

15:10 Improving Normalizing Flows to Sample from Boltzmann Distributions - Vincent Stimper

15:30 Panel discussion: - Chair: Lukas Heinrich
16:00 Coffee break
Coffee break
16:00 - 16:30
16:30 ML for phase transitions and sign problem mitigation: Contour deformation methods for noise reduction
Contour deformation methods for noise reduction
16:30 - 18:00
Contributions

16:30 Signal-to-noise improvement with contour deformations - Gurtej Kanwar

16:50 Complex normalizing flows and subtractions for sign problems - Yukari Yamauchi

17:10 Applying Complex Valued Neural Networks to the Hubbard Model Sign Problem: A Survey and Case Study - Marcel Rodekamp

17:30 Panel discussion: 1. maximum improvement in <sign> 2. ML towards such a maximal improvement 3. Contour deformation for specific observables 4. Translation between observables/theories - Chair: Will Detmold Members: Gurtej Kanwar Andrei Alexandru (The George Washington University) Thomas Luu Yukari Yamauchi
Tuesday, 28 February 2023
09:50 Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks - Frank Pollmann
Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks
- Frank Pollmann
09:50 - 10:20
10:20 Automatic differentiation for Lattice gauge theories - Alberto Ramos
Automatic differentiation for Lattice gauge theories
- Alberto Ramos
10:20 - 10:45
10:45 Coffee break
Coffee break
10:45 - 11:15
11:15 Gauge field generation
11:15 - 13:00
Contributions

11:15 Simulation of the 2D Schwinger Model via machine-learned flows in Global Correction steps - Jacob Finkenrath

11:40 Variational Autoregressive Networks for Information Theory - Tomasz Stebel

12:00 Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems - Jeanne Trinquier

12:20 Panel discussion: 1. Utility of ML for lattice 2. application-specific features of gauge field generation 3. Biggest challenges 4. Applications at an interesting scale - Chair: Phiala Shanahan Members: Fernando Romero Lopez Jacob Finkenrath Jeanne Trinquier
13:00 Group Photo
Group Photo
13:00 - 13:15

13:15 Lunch
Lunch
13:15 - 14:30
14:30 ML for phase transitions and sign problem mitigation
14:30 - 16:00
Contributions

14:30 Machine learning for quantum field theories with a sign problem - Andrei Alexandru

14:50 A machine learning approach to the classification of phase transitions in many flavor QCD - Marius Neumann

15:10 Teaching how to teach: defining learning samples to detect phase transitions - Francesco Di Renzo

15:30 Panel discussion: 1. Competitivity of ML for Phase identification 2. ML for Hypothesised Phases 3. Potential of ML for sign porblem - Chair: Maria Paola Lombardo
16:00 Coffee break
Coffee break
16:00 - 16:30
16:30 Spectral Reconstruction
16:30 - 18:00
Contributions

16:30 Brief introduction on spectral functions and their computation - Gert Aarts

16:40 Spectral reconstruction with Gaussian processes - Julian Urban

17:10 Spectral reconstruction with neural networks - Kai Zhou Lingxiao Wang

17:40 Learning regulators for spectral reconstruction - Alexander Rothkopf

17:50 Panel discussion - Chair: Gert Aarts
Wednesday, 1 March 2023
08:45 Introducing Munich Data Science Institute - Sylvia Kortüm
Introducing Munich Data Science Institute
- Sylvia Kortüm
08:45 - 09:00
09:00 EFT, Lattice and ML
09:00 - 10:40
Contributions

09:00 Learning Trivializing Gradient Flows - Simone Bacchio

09:40 An analysis of Bayesian estimates for missing higher orders in perturbative calculations - Aleksas Mazeliauskas

10:10 Generative models and EFTs - Marina Marinkovic
10:40 Coffee break
Coffee break
10:40 - 11:10
11:10 EFT, Lattice and ML
11:10 - 13:00
Contributions

11:10 Finite-volume pionless EFT for multi-nucleon systems with differential programming - Fernando Romero-Lopez

11:40 Panel discussion: 1. What can EFTs do for ML4LATTICE? 2. What ML4LATTICE could do for EFTs? - Chair: Nora Brambilla Members: Marina Marinkovic Phiala Shanahan William Detmold Aleksas Mazeliauskas
13:00 Lunch
Lunch
13:00 - 14:30
14:30 ML as component of exact algorithms
14:30 - 16:35
Contributions

14:30 Building Transport Maps and Making Good Use of Them - Michael Albergo

15:00 Automatic differentiation for Lattice QCD - A. Ramos

15:25 Symbolic Distillation of Neural Networks - Miles Cranmer

15:55 Panel discussion: 1. Meaning of "exact" algorithms 2. in-principle vs in-practice exactness 3. applications for exactness - Chair: Phiala Shanahan Members: Michael Albergo Andrei Alexandru Miles Cranmer
16:35 Reception
Reception
16:35 - 18:00
18:00 How to accelerate gauge field field generation using flow-based and hybrid models - Phiala Shanahan
How to accelerate gauge field field generation using flow-based and hybrid models
- Phiala Shanahan
18:00 - 19:00
Thursday, 2 March 2023
09:30 ML for particle physics - Lukas Heinrich
ML for particle physics
- Lukas Heinrich
09:30 - 10:00
10:00 Complex Langevin real-time simulations and ML - Alexander Rothkopf
Complex Langevin real-time simulations and ML
- Alexander Rothkopf
10:00 - 10:30
10:30 ML for physical interpretation of lattice results
10:30 - 10:50
Contributions

10:30 Towards fully bayesian analyses in Lattice QCD - Julien Frison
10:50 Coffee Break
Coffee Break
10:50 - 11:15
11:15 ML for physical interpretation of lattice results
11:15 - 13:00
Contributions

11:15 Physical Concepts from Neural Networks with Two Inputs - Sebastian Wetzel

11:35 Ab-Initio Quantum Chemistry via Graph Neural Networks - Nicholas Gao

12:00 Panel discussion on ML for physics interpretation - Chair: Alexander Rothkopf Members: Sebastian Wetzel Matteo Favoni Fernando Romero Lopez Miles Cranmer
13:00 Lunch
Lunch
13:00 - 14:30
14:30 Symmetry equivariant neural networks/informed neural networks
14:30 - 15:40
Contributions

14:30 Generation of gauge field configurations with equivariant neural networks - Matteo Favoni

14:50 TBA - Daniel Hackett

15:05 Panel discussion: 1. Symmetry-equivariance vs computational brute force 2. Computational challenges of gauge equivariance - Chair: Phiala Shanahan Members: Gurtej Kanwar Christoph Lehner Matteo Favoni
15:40 Poster session (With food and drinks)

15:40 - 18:40
Friday, 3 March 2023
09:00 ML and quantum field theories - Gert Aarts
ML and quantum field theories
- Gert Aarts
09:00 - 09:30
09:30 Inverse Problem
09:30 - 11:00
Contributions

09:30 Machine learning hadron spectral functions in Lattice QCD - Gabor Papp

10:00 R-ratio from Lattice QCD - Jian Liang

10:30 Inverse problem solving in nuclear physics with deep learning - Kai Zhou
11:00 Coffee break
Coffee break
11:00 - 11:30
11:30 Inverse Problem
11:30 - 12:30
Contributions

11:30 On the extraction of hadronic spectral densities from Euclidean correlators - Alessandro De Santis

12:00 Panel discussion - Chair: Andreas Kronfeld
12:30 Lunch
Lunch
12:30 - 14:00
14:00 EFT, Lattice and ML
14:00 - 15:15
Contributions

14:00 Machine learning for symbolic computation in high energy physics - Abdulhakim Alnuqayqdan

14:25 Alleviating the Sign Problem via Contour Deformation and Machine Learning - Thomas Luu

14:50 Gauge-equivariant neural networks as preconditioners in lattice QCD - Christoph Lehner
15:15 Coffee break
Coffee break
15:15 - 15:40
15:40 EFT, Lattice and ML
15:40 - 17:00
Contributions

15:40 Deep Learning and the Standard Model: a philosophy of science perspective - Luigi Scorzato

16:10 The route towards phase transition recognition in Lattice Gauge Theories - Andreas Athenodorou

16:40 Search for Exotic Higgs Bosons using Quantum Machine Learning - Nakul Aggarwal
17:00 Summary - Nora Brambilla
Summary
- Nora Brambilla
17:00 - 17:15

Choose timezone