Skip to content
View GVourvachakis's full-sized avatar
🎯
Focusing
🎯
Focusing
16 followers · 59 following

Achievements

Achievement: Pull Sharkx2Achievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Quickdraw

Block or report GVourvachakis

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
gvourvachakis/README.md

Hi, I'm Georgios 👋🏼

Research software for noisy data and stiff equations.

I'm Georgios Vourvachakis (Vakis), an independent researcher working across probabilistic machine learning, stochastic simulation, and numerical methods. I completed my MSc in Materials Science & Engineering (Computational Materials Science) at the University of Crete, with research at IACM and IESL, FORTH.

Selected work

  • TimeCausalVQVAE — Extending Time-Causal VAE baselines with causal VQ/RVQ tokenizers, autoregressive token priors, and path-risk diagnostics for market generation. Python package.
  • exponax-spde — Stochastic Allen–Cahn solvers built on Exponax, with exponential Euler–Maruyama integration, Q-Wiener noise, ensemble utilities, and numerical validation in JAX.

MSc research

Surrogate Modelling for Damage Threshold Estimation in Ultrashort Laser-Metal Interactions

Developed linear and nonlinear two-temperature-model solvers, including an IMEX Heun–Crank–Nicolson scheme with a strong Gauss–Seidel predictor, supported by stiffness and convergence studies. Built machine-learning surrogates for damage-threshold estimation.

Read the thesis · Thesis and presentation source

Tools

Python   JAX   PyTorch   NumPy   SciPy

I'm particularly interested in neural differential equations, stochastic flows, and generative modeling for financial time series.

Pinned Loading

  1. TimeCausalVQVAE TimeCausalVQVAE Public

    Time-causal financial generative models: refactored TC-VAE baselines with causal VQ/RVQ tokenizers, token priors, S&P500/VIX, Hawkes/SVMHJD, multi-dimensional benchmarks, and path-risk diagnostics.

    Python 1

  2. exponax-spde exponax-spde Public

    Stochastic PDE solvers (SPDE) built on top of exponax: Exponential Euler-Maruyama stepper for the stochastic Allen-Cahn equation with additive/multiplicative Q-Wiener noise, tamed nonlinearities, e…

    Python 7

  3. surrogate-damage-threshold-thesis surrogate-damage-threshold-thesis Public

    MSc thesis and presentation on surrogate modelling for damage-threshold estimation in ultrashort laser-metal interactions.

    TeX

  4. RDP_FedADMM RDP_FedADMM Public

    Privacy-preserving Federated ADMM with Rényi Differential Privacy, elastic weighted aggregation, and a statistical framework for non-IID data synthesis. Built on NumPy and sockets.

    Python

  5. CG_MD_of_Water_via_FM CG_MD_of_Water_via_FM Public

    A pipeline for coarse-graining (CG) molecular dynamics of homogeneous liquid water using the Force Matching (FM) method to reconstruct coarse-grained forces from fine-grained reference data.

    Python 1

  6. TMDs-Electronic-Structure-DFT TMDs-Electronic-Structure-DFT Public

    Systematic DFT study of electronic properties in Group-VI transition metal dichalcogenides (TMDs) using GPAW. Includes band structure calculations, density of states analysis, and orbital projectio…

    Python 1