My research revolves around interdisciplary applications of machine learning. The main focus of my PhD is applying ML to identify interesting phenomena arising in the plasma surrounding Earth at the magnetosphere.

Experience

  • Technical student at CERN working with Kubeflow, a cloud native service supporting scalable end-to-end machine learning pipelines
  • Research assistant at Helsinki Institute of Physics, calibration of particle jets at the CMS experiment
  • Full-stack developer at Fjuul devloping a mobile application for health tracking
  • Research assistant at Helsinki Accelerator Laboratory where we simulated iron-chromium alloys using molecular dynamics
  • Further information in my CV.

Publications

article Jet Energy Calibration with Deep Learning as a Kubeflow Pipeline
Daniel Holmberg, Dejan Golubovic, Henning Kirschenmann
Computing and Software for Big Science, Vol 7, Aug 2023
paper | arxiv | kubecon video | meetup video
article Jet Energy Corrections with Graph Neural Network Regression
Daniel Holmberg
Master’s Thesis, University of Helsinki, Feb 2022
thesis | slides | poster
article Interatomic Fe–Cr potential for modeling kinetics on Fe surfaces
Pekko Kuopanportti, Matti Ropo, Daniel Holmberg, et al.
Computational Materials Science, Vol 203, Feb 2022
paper | arxiv
article Potentialmodeller vid simulering av Fe-Cr
Daniel Holmberg
Bachelor’s Thesis, University of Helsinki, May 2019
thesis | slides

Talks

CERN ML Platform CMG Group Meeting :switzerland:
Jet Energy Corrections with GNN Regression using Kubeflow at CERN KubeCon Valencia :es:
Jet Energy Corrections with Graph Neural Network Regression Learning to Discover :fr:
Centralized Management of Your Machine Learning Lifecycle CERN IT Technical Forum :switzerland:
Jet Energy Corrections with DNN Regression CMS Machine Learning Forum :switzerland:
Top Quark Mass Measurement LPC CMS Data Analysis School
Atmospheric Aerosol Modeling HU Data Science Project :finland:
Emerging Computing Architectures HU Distributed ML Seminar :finland:
Interatomic Potentials for Fe-Cr Helsinki Accelerator Laboratory Simumeet :finland:

Workshops

Arnold Sommerfeld School: Physics meets Artificial Intelligence Ludwig Maximilian University
Plasma Physics Meets AI: Workshop on Subgrid-scale Modeling for Turbulence Aalto University
Representation Learning over Graph Data Institut Pascal
CMS ML Hackathon: Jet tagging CERN IdeaSquare
LPC CMS Data Analysis School Remote, Fermilab
CodeRefinery instructor training Remote, TU Delft
First online mega-CodeRefinery Online