QUC Summer School on “A.I. in High Energy Physics”



 July 4-15, 2022     

Rm 1503, Bldg 1, KIAS (Hybrid School)     

Program Home > Program


Rodrigo Alonso (IPPP, Durham)

Title: Higgs Effective Field Theory

Abstract: The nature of electro-weak symmetry breaking and the Higgs boson is one of the outstanding questions in particle physics and the answer is within reach in our era. Effective field theory (EFT) allows for the most general characterization of the Higgs in the absence of extra particles; we will identify in its formulation the possible fundamental characteristics of EWSB and the Higgs boson . Topics to be covered are the difference between the linear realization (aka SMEFT) and the general formulation (HEFT), ultra-violet completions, experimental prospects, and field theory of scalars in a curved manifold.

" We will follow the notes that can be found on  https://www.ippp.dur.ac.uk/~ralonso/PnlWPG.html .  Exercises for students to try can be found between horizontal teal-colored lines, the solutions will be provided later. ''



Cliff Burgess (MacMaster and Perimeter)

Title: Thinking Effectively About Gravity and Cosmology


We live at a time of contradictory messages about how successfully we understand gravity in general and cosmology in particular. General Relativity seems to work well in the Earth’s immediate neighbourhood, but arguments abound that it needs modification at very small and/or very large distances. Lambda-CDM cosmology seems to provide a precision understanding of the Universe we see around ourselves, but only at the expense of having to introduce *two* new types of matter (Dark Matter and Dark Energy). Gravitational evolution of quantum fluctuations in the early Universe are argued to provide good descriptions of primordial matter distributions but reconciling quantum mechanics and gravity is a well-known long-standing unsolved conceptual problem. How should evidence be assessed; which arguments are speculative and which can be trusted? These lectures try to assess theoretical uncertainties by embedding them into the broader context of similar situations in other areas of physics. This is done by emphasizing how Effective Field Theories underpin our quantitative understanding of theoretical error. The main message is mixed: On one hand cosmology seems to like features (like light scalars and small vacuum energies) that are not generic to the long-wavelength limit of fundamental theories, and this is a crucial clue that would be silly to ignore. On the other hand, although some ways are known to contrive these properties (and so among which we can seek explanations for observations), so far none are known that all researchers agree describe all observations (even in principle), making this a clue that seems difficult to use (at least until a convincing example is found).


Clifford Cheung (Cal Tech, EFT, S-matrix)

Title: Introduction to Scattering Amplitudes

Abstract: Scattering theory is a powerful framework for understanding the fundamental laws of nature.  In this approach one relinquishes the standard tools of quantum field theory, instead bootstrapping scattering amplitudes directly from physical principles alone.  Through this process one can then derive classic results from field theory as well as new, far more mysterious structures residing within the theories that describe our world, including the strong interactions and Einstein's general relativity.  In these introductory lectures I discuss the modern scattering amplitudes program, focusing on i) the principles and implementation of the bootstrap, as well as ii) the novel amplitudes structures that connect and unify gravity, gauge theory, and effective field theories and their lessons for quantum field theory.


Yu-tin Huang (National Taiwan Univ)

Title: Amplitudes

Lecture plans: 

2-hr introduction to amplitudes focusing on on-shell variables and applied to three-particle interaction and constraints on 2 to 2 scattering. 

2-hr discussion on dispersion relations for the 2-2 scattering, and the extraction of constraints on general effective field theories including gravity. 

2-hr discussion on extracting the two body classical dynamics from 2-2 scattering amplitude, focusing on binary black hole dynamics. 

2-hr  Advanced topics: TBD can cover on-shell approach to anomalies, monopole scattering, or geometric Amplitudehedron for N=4 SYM and ABJM. 



Isaac Kim (UC Davis)

Lecture 1: Basics of Quantum Computing
Abstract: In this lecture, I will discuss the brief history of quantum computing, focusing on the important recent theoretical and experimental developments. Then I will discuss the basics, e.g., qubit, Hilbert space, and quantum gates.

Lecture 2: Quantum Circuit
Abstract: I will introduce the quantum circuit model and discuss various useful facts that are used in the literature. 

Lecture 3: Hamiltonian Simulation Algorithms: Part 1
Abstract: I will explain one of the main application of quantum computing: simulating physical systems of interest. I will explain what Hamiltonian simulation precisely is and discuss its applications, e.g., quantum phase estimation.

Lecture 4: Hamiltonian Simulation Algorithms: Part 2
Abstract: I will discuss modern quantum algorithms for Hamiltonian simulation, e.g., linear combination of unitaries and qubitization.


Alex Pomarol (Barcelona, CERN and KIAS)

Title : Amplitudes for EFT

Abstract : I will start describing how effective theories (EFT) can be characterized by on-shell amplitudes (instead of Lagrangians of fields). Later, I will show how amplitude methods allow to derive one-loop renormalization effects from just tree-level on-shell amplitudes, with no need of loop calculations. 


David Shih (Rutgers Univ), Gregor Kasieczka (Univ. of Hamburg, Machine Learning)

Title: "Modern Machine Learning and Particle Physics" 

Abstract: Deep learning and artificial intelligence are revolutionizing nearly every corner of science, engineering and beyond. We will give a pedagogical introduction to the basics of modern machine learning and some of its recent exciting applications to particle physics. Topics we will cover include neural networks, gradient descent, backpropagation, classification, regression, generative modeling, and anomaly detection.


Lecture 1    David       Motivations / ML basics                                                                  
Lecture 2    David       Classifiers (with top tagging as an example)                                       
Lecture 3    Gregor     Data representations and architectures (top tagging landscape)        
Lecture 4    Gregor     Decorrelation+uncertainties                               
Lecture 5    Gregor     Generative models 1                                       
Lecture 6    David       Generative models 2                                      
Lecture 7    Gregor     Anomaly detection 1                                       
Lecture 8    David       Anomaly detection 2                                      

Tutorial 1   Sung Hak   Basics of ML, gradients, simple examples
Tutorial 2   Sung Hak   Classifier examples with top tagging dataset
Tutorial 3   Sung Hak   Normalizing flows with example(s)



Michael Trott (Niels Bohr Institute)

Title: The SMEFT/geoSMEFT- the modern EFT analysis of collider data

Abstract: I will define the modern Standard Model Effective Field Theory (SMEFT);
the theoretical paradigm for interpreting LHC and lower energy data (below a cut off scale Lambda).
The key aspects of the phenomenology of the SMEFT, that differ from beyond the standard model (model building)
assumptions and approaches, will be discussed. The recent W mass measurement anomaly will be used as an example
of the utility of SMEFT analyses. The SMEFT is profitably thought of as a geometric field theory for high
order calculations in the 1/Lambda expansion defining the theory - this defines the geoSMEFT.
This very recent field theory construction will be defined and its utility demonstrated in some some simple examples.




Ben Nachman (LBNL, UC Berkeley)

Title: Discovering Unanticipated New Physics With Machine Learning
Abstract: Modern machine learning tools are allowing us to explore high energy physics data in new ways. These approaches may enable discoveries that were unthinkable with existing strategies. I introduce these new “anomaly detection” methods, which differ from typical approaches developed for non-scientific applications. While these methods are just starting to be applied to experimental data, there is an exciting program ahead of us as these tools become more developed and widely deployed.



Jesse Thaler (MIT/IAIFI)
Title: Weak Supervision for the Strong Force

Abstract: The strong nuclear force is governed by the interactions of quarks and gluons. Because of confinement, though, quarks and gluons can never been seen in isolation, so “quark” and “gluon” labels are fundamentally ambiguous. In this talk, I show how to leverage weak supervision to disentangle quarks and gluons without labeled training data. This technique is then applied to public data from the Large Hadron Collider. This analysis incorporates a wide range of machine learning tools — including topic modeling, permutation-invariant networks, simulation-based inference, and optimal transport — together with key insights from quantum field theory.



Max Tegmark (MIT/ IAIFI/CBMM)
Title: AI for physics & physics for AI

Bio: Max Tegmark is a professor doing AI and physics research at MIT as part of the Institute for Artificial Intelligence & Fundamental Interactions and the Center for Brains, Minds and Machines. He advocates for positive use of technology as president of the Future of Life Institute. He is the author of over 250 publications as well as the New York Times bestsellers “Life 3.0: Being Human in the Age of Artificial Intelligence” and "Our Mathematical Universe: My Quest for the Ultimate Nature of Reality". His most recent AI research focuses on intelligible intelligence.