CV •

Curriculum Vitae

I am a scientist with 12+ years experience applying quantitative methods to solve challenging research problems in physics and national security. I have 7+ years experience in technical leadership roles and as a principal investigator leading my own research team, with an established record of high-impact publications, presentations at major international conferences and novel applications of machine learning to physics problems.

Education

Columbia University, New York, NY

• Ph.D. in Physics, 2012
• M.Phil., 2009; M.A., 2008
• B.S. Applied Physics, 2006, magna cum laude

Professional Experience

Lawrence Livermore National Laboratory, Livermore, CA

Staff Physicist, 2017–2024

Leader of Heavy Ion Physics Group

• Principal investigator on 6 research projects in particle and nuclear physics
• Founded group and managed a team of 2–3 physicists, a data scientist and several student interns

Convener of Heavy Ion Group in the ATLAS Collaboration at the LHC (2022–2024)

• Coordinated activities of approximately 50 students, postdocs and senior physicists to accomplish data acquisition, processing, analysis, interpretation and publication of scientific results
• Oversaw submission of 20+ publications

Columbia University, New York, NY

Postdoctoral Research Scientist, 2012–2017

• Designed and implemented industry-leading algorithm for jet reconstruction in heavy-ion collisions
• Established methodology for data-driven uncertainty quantification now widely used in the field
• Led analysis and publication of several high-impact results in areas of relativistic heavy-ion collisions and presented results at major international conferences

Selected Machine Learning Projects

Particle Reconstruction

Developed CNNs and graph neural networks to reconstruct particles using multi-modal signals from particle detectors using different technologies. Achieved 50% improvement in energy resolution and reduced false positive rate by 10x while maintaining >95% recall in classifying particle type. Publications: CNNs, Graph Neural Networks

AI-Assisted Detector Design

Cast design of future particle detectors as a formal optimization problem, improving energy resolution by 40% over baseline designs while establishing a new tool for choosing design parameters. Publications: Segmentation Optimization, SiPM-on-Tile ZDC

Generative Modeling

Developed generative models as fast surrogates for detector simulation utilizing diffusion models to generate point-cloud and image representations of calorimeter showers with minimal fidelity loss. Point-cloud models achieve better performance with 4x faster generation time. Publication: Point Cloud vs Image Models

Neural Simulation-Based Inference

Utilized DNNs to represent high-dimensional likelihood ratios for parameter inference in physics models, reducing Higgs boson self-coupling parameter confidence intervals by 2–3x.

Skills and Qualifications

Scientific

Strong analytical reasoning skills, physics insight and experimental methodology; deep knowledge of mathematical models and statistical methods; applied machine learning

Data and Statistical Analysis

Extensive experience with large datasets from data acquisition, analysis, uncertainty quantification and statistical inference

Computational

C++, Python, familiarity with ML frameworks (scikit-learn, PyTorch) and ecosystem; production environment/workflow (Dev/GitOps), HPC and grid computing; mathematical/physics modeling, development of novel pattern recognition algorithms

Technical Communication

Strong scientific writing and presentation skills; referee for JHEP, PLB and PRL, and reviewer for DOE Office of Science proposals; presentations at international conferences

Selected Publications

For complete publication list, see the dedicated Publications page.

Machine Learning Applications

• R. Milton et al., Design of a SiPM-on-Tile ZDC for the future EIC and its Performance with Graph Neural Networks, Submitted to JINST (2024), arXiv:2406.12877

• F. Torales Acosta et al., The Optimal use of Segmentation for Sampling Calorimeters, JINST 19 (2024) 06, P06002, arXiv:2310.04442

• F. Torales Acosta et al., Comparison of Point Cloud and Image-based Models for Calorimeter Fast Simulation, JINST 19 (2024) 05, P05003, arXiv:2307.04780

• ATLAS Collaboration, Point Cloud Deep Learning Methods for Pion Reconstruction in the ATLAS Experiment, ATL-PHYS-PUB-2022-040 (2022), CDS Record

• ATLAS Collaboration, Deep Learning for Pion Identification and Energy Calibration with the ATLAS Detector, ATL-PHYS-PUB-2020-018 (2020), CDS Record

Landmark Heavy-Ion Physics Results

• ATLAS Collaboration, Observation of a Centrality-Dependent Dijet Asymmetry in Lead–Lead Collisions at √sNN= 2.76 TeV with the ATLAS Detector at the LHC, Phys. Rev. Lett. 105 (2010) 252303, arXiv:1011.6182 (First jet measurement in heavy-ion collisions)

• ATLAS Collaboration, Measurement of photonuclear jet production in ultra-peripheral Pb+Pb collisions at √sNN=5.02 TeV with the ATLAS detector, Submitted to Phys. Rev. D. (2024), arXiv:2409.11060

• ATLAS Collaboration, Observation of Light-by-Light Scattering in Ultraperipheral Pb+Pb collisions with the ATLAS Detector, Phys. Rev. Lett. 123 (2019) 052001, arXiv:1904.03536

Recent Invited Talks

For complete presentation history, see the dedicated Talks page.

Major International Conferences

• ATLAS Overview, The XXVIII International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions (“Quark Matter”), Houston, TX, September 2023

• Ultra-peripheral collisions, The XXVII International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions (“Quark Matter”), Venice, Italy, May 2018

• ATLAS highlights, The XXV International Conference on Ultra-Relativistic Nucleus-Nucleus Collisions (“Quark Matter”), Kobe, Japan, September 2015

Recent Lectures & Seminars

• Applications of AI and ML to Nuclear Physics, Lectures at 2023 National Nuclear Physics Summer School, Riverside, CA, July 2023

• Machine Learning Approaches to Calorimetric Particle Reconstruction, University of Washington, August 2022

Invited seminars at: BNL (2013), CERN (2014), Colorado (2016), Columbia (2013, 2016), LBNL (2018), LLNL (2017), Ohio (2020), Penn State (2018), SLAC (2014)

Service and Leadership

• Scientific coordinator for ATLAS heavy-ion physics program
• Convener and organizer of international conference sessions on heavy-ion physics
• Reviewer for major physics journals including Physical Review Letters, Journal of High Energy Physics

Technical Skills

• Programming: Python, C++, ROOT, machine learning frameworks (TensorFlow, PyTorch)
• Statistical Analysis: Advanced statistical methods, uncertainty quantification, Monte Carlo techniques
• Experimental Physics: Large-scale detector systems, data acquisition, trigger systems
• Leadership: International collaboration management, scientific project coordination

  Doctoral Thesis

  Jet Quenching in Relativistic Heavy Ion Collisions at the LHC
  Columbia University, 2012

• ATLAS Thesis Award - Recognizing outstanding contributions to ATLAS in the context of Ph.D. research
• Springer Thesis Award - Selected for publication via Springer Theses
• Manuscript available on arXiv and CDS