Tracking with Graph Neural Networks

High Energy Physics Machine Learning

Real-time particle track reconstruction from high-frequency, noisy data using Geometric Deep Learning, namely Graph Neural Networks (GNNs), on GPUs and FPGAs at the LHC.

etx4velo-1 preview Figure: Illustration of the process of moving from hits in the detector to a graph of the event. etx4velo-2 preview Figure: Illustration of the process of moving from the event graph to the reconstructed tracks of the event.

Paper, Paper, Poster, Poster, Code(Python, C++, CUDA)

Supervision: Vava Gligorov, Bertrand Granado, Vladimir Lončar

Traffic Anomaly Detection

Machine Learning Anomaly Detection

Learning traffic anomalies from generative models on real-time spatio-temporal data from 125 cameras in Gothenburg, Sweden.

xim-1 xim-2 Figure: Detection of the beginning of heavy snowfall. Scenes from Camera 14 on Nov. 19, 2020, at 14:10 (left) and 14:20 (right).

Paper, Presentation, Poster, Code(Python)

Supervision: Alexandros Sopasakis

Energy Transport in the PXP Spin Chain

Quantum Chaos Many-Body Physics

Exploring weak ergodicity breaking in the PXP spin chain.

pxp-1 Figure: Spins on a chain, interacting only if they are close to each other. Figure by Alexander Wagner. pxp-2 Figure: Observation of linear fronts in the space-time diagram of the evolution of the energy density of a PXP chain.

Master Thesis, Code(Julia, Python)

Supervision: Spyros Sotiriadis, Achileas Lazaridis

Chimera States in Oscillator Networks

Nonlinear Dynamics Complex Systems

Chimera states in the leaky integrate-and-fire model of spiking neuron oscillators.

lif-1 Figure: Development of a chimera state in a network of coupled identical oscillators, with coexisting domains of coherence and incoherence.

Project Report, Code(Python, C++, Java)

Supervision: Astero Provata

Exploring Trading Strategies for Crypto

Quantitative Finance Cryptocurrencies

xbtusd-1 Figure: Illustration of buy and sell signals based on moving averages, on historical data of bitcoin prices.

Code(Python), Code(Python)

Yang–Mills Existence and Mass Gap

Theoretical Physics Mathematical Physics

Studied the “Yang–Mills existence and mass gap problem”, an open question in mathematical physics and mathematics, and one of the seven Millennium Prize Problems established by the Clay Mathematics Institute, which offers a $1M reward for a correct solution.

\[ \mathcal{L} = -\frac{1}{2} \text{tr}(F^2) = -\frac{1}{4} F_{\mu\nu}^a F^{a\mu\nu} \]

Equation: Lagrangian of gauge theories with a non-abelian symmetry group.

\[ \mathcal{L}_{\text{SM}} = \mathcal{L}_{\text{gauge}} + \mathcal{L}_{\text{fermion}} + \mathcal{L}_{\text{Higgs}} + \mathcal{L}_{\text{Yukawa}} \] \[ \begin{aligned} \mathcal{L}_{\text{gauge}} &= -\frac{1}{4}G_{\mu\nu}^a G^{a\mu\nu} - \frac{1}{4}W_{\mu\nu}^i W^{i\mu\nu} - \frac{1}{4}B_{\mu\nu} B^{\mu\nu} \\ \mathcal{L}_{\text{fermion}} &= \sum_{\psi} \bar{\psi} i\gamma^\mu D_\mu \psi \\ \mathcal{L}_{\text{Higgs}} &= (D_\mu \phi)^\dagger(D^\mu \phi) - V(\phi), \quad V(\phi) = \mu^2 \phi^\dagger \phi + \lambda (\phi^\dagger \phi)^2 \\ \mathcal{L}_{\text{Yukawa}} &= -\left( y_u \bar{Q}_L \tilde{\phi} u_R + y_d \bar{Q}_L \phi d_R + y_e \bar{L}_L \phi e_R + \text{h.c.} \right) \end{aligned} \]

Equations: The Standard Model Lagrangian \(\mathcal{L}_{\text{SM}}\), showing gauge, fermion, Higgs, and Yukawa sectors.

Master Thesis

Supervision: Philip Candelas, Christopher Beem

Spectra of PAHs Found in Space

Astronomy Astrophysics Computational Chemistry

Investigating the spectra of Polycyclic Aromatic Hydrocarbons (PAHs) found in space. Their electronic structure is modeled using density functional theory.

xbtusd-1 Figure: Calculation of the infrared spectrum of \( C_{52}H_{18}N^{++} \) using the PM3 semi-empirical method.

Project Report

Supervision: Dimitra Rigopolou, Patrick Roche

Other Resources

Physics Astronomy