Real-time decision making at 40 MHz at the Large Hadron Collider

The UKRI AI Centre for Doctoral Training (CDT) in Decision Making for Complex Systems is a joint CDT between The University of Manchester and the University of Cambridge. The CDT provides funding for four years of advanced studies towards a PhD. The first year consists of a taught program at Manchester that will cover the fundamentals of Machine Learning. This year is followed by three years of research at either at Manchester or Cambridge. Please note the research element of the PhD will take place at the host institution of the supervisor listed for each project.

The Standard Model of particle physics describes all known fundamental particles and their non-gravitational interactions, but it lacks any particle consistent with the astrophysical evidence for dark matter.

The student working on this PhD project will use novel data taking and machine learning techniques to gain insight on the particle nature of dark matter, using data from the ATLAS experiment at the Large Hadron Collider. The student will play a leading role in recording new datasets using the real-time decision making system of the ATLAS detector (the “trigger system”), and use them to search for yet unexplored dark matter hypotheses at the High Luminosity LHC (2029 onwards) and in future colliders.

The student will significantly enhance the ATLAS trigger system, boosting the experiment’s overall discovery potential, especially in dark matter searches. This involves real-time analysis, departing from the conventional process of first collecting data and then analysing it. Leveraging machine learning, specifically unsupervised outlier detection algorithms, allows for more efficient analysis of vast amounts of data, retaining only essential information for outlier events. This approach enhances sensitivity to unexplored dark matter particles, even those not yet theorised.

The PhD student will work to meet the main challenges in using these methods on a highly complex dataset using resource-constrained computing environments: 

1)How well do outlier detection algorithms perform in environments with “unknown unknowns,” like detector malfunctions (noise)? Can they differentiate between various types of outliers?

2)Do we understand the decision-making process of the algorithm enough so that it is interpretable and reproducible?  

3)Can we reduce the energetic footprint of the algorithm in light of the upcoming LHC upgrade which will deliver 100x more data to the ATLAS detector (for 40 MHz real-time analysis), and towards future colliders where energetic considerations will be key for the sustainability of the whole project? 

This PhD project will be fully integrated in the European Training Network SMARTHEP (www.smarthep.org), coordinated by the PhD supervisor. The student will collaborate with industrial partners such as IBM and Verizon (note: we are currently hosting a Verizon Early Stage Researcher working on related topics) and discuss complementary ML techniques to outlier detection within IBM expertise such as rule induction.

The outcomes of this project will be:

a)Innovative data-taking and real-time analysis methods and outlier detection algorithms, implemented in the ATLAS trigger system but also available off-the-shelf for other scientists to adapt and use for their use case;

b)A discovery, or world-leading constraints on not yet probed kinds of dark matter.

The results obtained will be disseminated as peer-reviewed papers and conference talks, as well as publicly available tools within Manchester-led initiatives such as the European DM collaboration between experimental and theory groups (iDMEu) and the European Science Cluster of Astronomy and Particle Physics ESFRI research infrastructures (ESCAPE) Dark Matter Science project. This will ensure that data, algorithms and results are fully compliant with FAIR principles of Open Science. The software written and used for this project will meet the quality criteria set out by the new EU-funded European Virtual Institute for Research Software Excellence (EVERSE) which sees leading actors in Physics & Astronomy and Computer Science at the University of Manchester, including the main supervisor as Work Package leader. The presence of other scientific infrastructures (e.g. SKAO) in both CDT and in these projects will facilitate cross-talk and interdisciplinary collaboration between this and other student projects and for the benefit of the whole CDT. 

Entry requirements

Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master’s (or international equivalent) in a relevant science or engineering related discipline.

How to Apply

As the CDT has only recently been awarded we strongly encourage you to contact the supervisor of the project you are interested in with your CV and supporting documents. You will have a chance to meet with prospective supervisors prior to submitting an application – further details will be provided.

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact.

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

We also support applications from those returning from a career break or other roles. We consider offering flexible study arrangements (including part-time: 50%, 60% or 80%, depending on the project/funder).

Supporting documents required

We will conduct a pre-screening interview for selected candidates (official paperwork will be requested later). To be considered in the selection of interviewees, please send us:

(a) A CV, including your GPA

(b) A research statement of maximum 2 pages font size 11, explaining your background and interest in the field. You should also isolate three questions related to the research project, motivate their importance and suggest how your contribution will bring us closer to answering them

(c) Transcripts

Please specify in your CV and/or research statement your skills in the following:

(i) Undergraduate research experience in particle physics

(ii) Documented proficiency in Python and C++ (e.g. GitHub repository with example codes)

(iii) Proficiency in mathematical methods such as those needed for the Quantum Field Theory or Mathematical Topics in Machine Learning courses at the University of Manchester (see below)

(iv) Experience in machine learning, if any

^{AI_CDT_DecisionMaking}