Machine Learning at GGI - Conference

5 Sept 2022, 09:00 → 9 Sept 2022, 16:45 Europe/Rome

 

Abstract
Machine learning (ML) is nowadays an important toolbox for theoretical and experimental physics, and its importance is expected to steadily grow in the coming years. Thanks to its effectiveness and extreme flexibility, it allows for applications covering a huge set of topics, ranging from statistical data analysis, to simulation and modeling. For this reason ML has been successfully used in very different research areas, such as high-energy physics, astrophysics and cosmology, condensed matter and statistical physics.

Applications in different domains often share strong similarities either in the problems to be solved or in the methodology employed. This motivates a fruitful exchange of ideas, which however is seldom achieved in practice due to the distance among different research communities.

The aim of the workshop is to bring together researchers with interests and expertise in ML from different fields in physics, strongly encouraging and promoting cross-topic exchange of ideas and collaborations. Three broad research areas will be covered:
- High-Energy Physics
- Astrophysics, Cosmology and Astroparticles
- Condensed Matter and Statistical Physics (including Quantum Information)

The distinctive trait of the workshop will be the focus on theoretical physics in a broad sense, including data analysis as well as simulation and modelling.

Topics
- Methods for regression and statistical analysis
- Monte Carlo integration and simulation
- Anomaly detection
- Classification
- Time series analysis
- Clustering and multi-dimensional visualization
- Equation solving
- Artificial intelligence-inspired and -augmented science
- Statistical physics algorithms for optimization and learning problems
- Quantum machine learning


Organizers
Massimo Brescia (INAF Napoli)
Filippo Caruso (U. Firenze)
S. George Djorgovski (Caltech)
Duccio Fanelli (U. Firenze)
Alessandro Marconi (U. Firenze)
Florian Marquardt (Max Planck Erlangen)
Giuliano Panico (U. Firenze)
Jesse Thaler (MIT)
Andrea Wulzer (CERN & U. Padova)


Local organizer
Giuliano Panico (U. Firenze)

Contact
giuliano.panico@unifi.it

Monday, 5 September

09:00 → 10:30 Registration

10:30 → 11:00 Coffee break

11:00 → 11:15 Welcome

11:15 → 13:00 Condensed Matter/QI

11:15 Machine Learning and Large Scale Quantum Systems

Speaker: Eliska Greplova

11:50 Machine Learning for quantum noise discrimination

Speaker: Stefano Martina

slides_martina.pdf

12:25 Spectral Pruning for Neural Network

Speaker: Lorenzo Giambagli

13:00 → 14:30 Lunch

14:30 → 15:40 Astro/Cosmo

14:30 A semi-supervised approach to dark matter searches in direct detection data with machine learning

Speaker: Andre Scaffidi

GGI_talk_Andre_Scaffidi.pdf

15:05 Two machine learning based methods for field-level inference of cosmological parameters

Speaker: Moritz Münchmeyer

GGI munchmeyer - new techniques for field-level ML in cosmology.pdf

15:40 → 16:15 Astro/Cosmo

15:40 Recurrent NNs for fast signal discovery in pulsar timing arrays

Speaker: Marat Freytsis

16:15 → 16:45 Coffee break

16:45 → 17:55 HEP

16:45 Using Machine Learning for Physics: The Underlying Mathematical Assumptions

Speaker: Jeff Byers

Byers_Conf_Talk_GGI_Sep5.pptx

17:20 Product Jacobi-Theta Boltzmann machines with score matching

Speaker: Andrea Pasquale

talk_Pasquale.pdf

Tuesday, 6 September

10:10 → 10:40 Coffee break

10:40 → 13:00 Condensed Matter/QI

10:40 Quantum Computing & Quantum Machine Learning

Speaker: Filippo Caruso

11:15 Algorithmic exploitation of the energy landscape in non-convex neural networks

Speaker: Riccardo Zecchina

11:50 Coarse-graining for classical and quantum systems

Speaker: Cecilia Clementi

12:25 Bridge between Classical & Quantum Machine Learning

Speaker: Jack Araz

jackaraz_ggi.pdf

13:00 → 14:30 Lunch

14:30 → 16:15 Astro/Cosmo

14:30 Cosmological simulations of Galaxy formation supervised by Machine Learning algorithms

Speaker: Luca Graziani

15:05 Reconstructing blended galaxies with Machine Learning

Speaker: Lavanya Nemani

GGI Presentation_Nemani.pdf

15:40 Digging into data cubes with Deep Learning

Speaker: Giuseppe Longo

ML@GGI-Longo.pdf

16:15 → 16:45 Coffee break

16:45 → 17:55 HEP

16:45 Style-based quantum generative adversarial networks for Monte Carlo events

Speaker: Stefano Carrazza

17:20 PDF determination: from NN fitting to posterior sampling

Speaker: Alessandro Candido

talk_Candido.pdf

Wednesday, 7 September

10:10 → 10:40 Coffee break

10:40 → 13:00 HEP

10:40 The LHC Physics case

Speakers: Marco Zanetti (Istituto Nazionale di Fisica Nucleare), Marco Zanetti (CERN)

AnomaliesAtLHC_GGI.pptx

11:15 Learning New Physics from a Machine

Speaker: Gaia Grosso

ML_GGI_conference_GGrosso.pdf

11:50 Efficient nonparametric methods for statistical anomaly detection

Speaker: Marco Letizia

12:25 Challenges for unsupervised anomaly detection in particle physics

Speaker: Katherine Fraser

Fraser_GGI_2022_Talk.pdf

13:00 → 14:30 Lunch

14:30 → 16:15 Condensed Matter/QI

14:30 A theory explaining the limits and performances of algorithms based on simulated annealing in solving sparse hard inference problems

Speaker: Federico Ricci Tersenghi

15:05 Quantum associative memory for quantum image recognition

Speaker: Sreetama Das

15:40 Reinforcement Learning for Quantum Technologies

Speaker: Florian Marquardt

16:15 → 16:45 Coffee break

16:45 → 17:20 Astro/Cosmo

16:45 Applying DNN to extract nuclear matter properties from neutron stars

Speaker: Marcio Ferreira

Talk_Ferreira_GGI2022.pdf

18:00 → 20:30 Visit to Galileo's Villa il Gioiello

Thursday, 8 September

10:10 → 10:40 Coffee break

10:40 → 13:00 Astro/Cosmo

10:40 NCPA mitigation using PCA and Neural Network

Speaker: Alessandro Terreri

11:15 Accelerating cosmological inference with Deep Learning

Speaker: Marco Bonici

11:50 Graph-Convolutional Neural Networks as a new tool to extract clustering information of the Cosmic web

Speaker: Farida Farsian

12:25 Optimal compression of the cosmic 21-cm signal

Speaker: David Prelogovic

13:00 → 14:30 Lunch

14:30 → 16:15 HEP

14:30 Parametrized classifiers for optimal EFT sensitivity

Speaker: Giuliano Panico

Panico_ML_precision.pdf

15:05 Towards an optimal global SMEFT fit with machine learning

Speaker: Jaco Ter Hoeve

GGI_terHoeve.pdf

15:40 Exploring EFT with ML at the LHC

Speaker: Robert Schöfbeck

22-08-19 GGI workshop - ML in SM-EFT - Conference.pdf

16:15 → 16:45 Coffee break

16:45 → 17:55 HEP

16:45 PHYSTAT Activities

Speaker: Louis Lyons

GGI_PHYSTAT.pptx

17:20 Unsupervised Quark/Gluon Jet Tagging With Poissonian Mixture Models

Speaker: Manuel Szewc

Conference_GGI_Szewc.pdf

Friday, 9 September

10:10 → 10:40 Coffee break

10:40 → 11:15 Astro/Cosmo

10:40 Translation and Rotation Equivariant Normalizing Flows for cosmology and beyond

Speaker: Uros Seljak

11:15 → 13:00 HEP

11:15 Particle Cloud Generation with Message Passing GANs

Speaker: Raghav Kansal

ML_GGI_09_09_22.pdf

11:50 Gradient estimators for normalising flows

Speaker: Piotr Bialas

12:25 Machine learning for supersymmetric matrix models

Speaker: Enrico Rinaldi

13:00 → 14:30 Lunch

14:30 → 16:15 Condensed Matter/QI

14:30 Rapid Exploration of Topological Band Structures Using Deep Learning

Speaker: Vittorio Peano

15:05 Renormalized Mutual Information for Artificial Scientific Discovery

Speaker: Leopoldo Sarra

15:40 Quantum density peak clustering

Speaker: Lorenzo Buffoni

16:15 → 16:45 Coffee break