Gorka Muñoz-Gil

GorkaMuñoz-Gil

Machine Learning for Science · University of Innsbruck

My research focuses on developing new machine learning algorithms and using them to discover new physics phenomena in the quantum and classical realm.

CV (pdf) Google Scholar ↗ ORCID ↗ GitHub ↗
Gorka Muñoz-Gil
Randomness in Physics

Research

ML in the quantum domain

In recent years, my main focus has been using ML to push the boundaries of quantum science. One research line explores generative AI, in particular diffusion denoising models, for generating quantum circuits. Our proposal was featured on the cover of Nature Machine Intelligence. We are now expanding this work on several fronts, especially through a collaboration with Nvidia, leveraging their new CUDA-Q library and hardware (see our new publication).

Another key direction of my work aims to develop interpretable ML approaches for discovering new quantum phenomena. We recently demonstrated how variational autoencoders can uncover the phase structure of experimental quantum simulators—independently of their implementation—from Rydberg atom arrays in Prof. Lukin’s group at Harvard to Bose–Einstein condensates in Prof. Schmiedmayer’s group in Vienna.

ML in soft and active matter

Anomalous diffusion was the focus of my PhD thesis. In collaboration with the biophysics groups of Prof. Manzo (Vic) and Prof. Volpe (Gothenburg), we pioneered the use of ML to characterize diffusion trajectories through the AnDi Challenge, a global scientific competition with over 150 participants across two editions. The results of the latest challenge were recently published in Nature Communications.

More recently, I have explored optimal search strategies using interpretable reinforcement learning, gaining theoretical insights applicable to the motion of intelligent systems across soft matter and ecology (see paper, and others in the Publications tab).

Science beyond science

Science doesn’t end with publications. I enjoy engaging in projects that extend my involvement in science beyond the lab—from teaching to outreach. I currently teach the Machine Learning in Physics Master course and have led a summer camp on quantum information for high school students (see Teaching tab).

I am also committed to open science and outreach, from citizen-science initiatives such as NightUp! to public talks at Pint of Science and artistic collaborations—like a quantum-inspired music piece performed by Reiko Yamada at the Sonar Festival 2021.

News

Selected papers

  1. Quantum circuit synthesis with diffusion models
    Fürrutter, Florian, Muñoz-Gil, Gorka and Briegel, Hans J
    Nat. Mach. Intell. Nature Machine Intelligence · 2024
    arXiv link

    Quantum computing has recently emerged as a transformative technology. Yet, its promised advantages rely on efficiently translating quantum operations into viable physical realizations. In this work, we use generative machine learning models, specifically denoising diffusion models (DMs), to facilitate this transformation. Leveraging text-conditioning, we steer the model to produce desired quantum operations within gate-based quantum circuits. Notably, DMs allow to sidestep during training the exponential overhead inherent in the classical simulation of quantum dynamics – a consistent bottleneck in preceding ML techniques. We demonstrate the model’s capabilities across two tasks: entanglement generation and unitary compilation. The model excels at generating new circuits and supports typical DM extensions such as masking and editing to, for instance, align the circuit generation to the constraints of the targeted quantum device. Given their flexibility and generalization abilities, we envision DMs as pivotal in quantum circuit synthesis, enhancing both practical applications but also insights into theoretical quantum computation.

    @article{furrutter2023quantum,
  title = {Quantum circuit synthesis with diffusion models},
  author = {F{\"u}rrutter, Florian and Mu{\~n}oz-Gil, Gorka and Briegel, Hans J},
  journal = {Nature Machine Intelligence},
  pages = {1--10},
  year = {2024},
  arxiv = {2311.02041},
  paper_image = {gen_qc.png},
  bibtex_show = {true},
  abbr = {Nat. Mach. Intell.},
  html = {https://doi.org/10.1038/s42256-024-00831-9},
  selected = {true}
}
  2. Quantitative evaluation of methods to analyze motion changes in single-particle experiments (AnDi Challenge 2)
    Muñoz-Gil, Gorka, Bachimanchi, Harshith, Pineda, Jesús [...], Volpe, Giovanni and Manzo, Carlo
    Nat. Commun. Nature Communications · 2025
    arXiv link

    The analysis of live-cell single-molecule imaging experiments can reveal valuable information about the heterogeneity of transport processes and interactions between cell components. These characteristics are seen as motion changes in the particle trajectories. Despite the existence of multiple approaches to carry out this type of analysis, no objective assessment of these methods has been performed so far. Here, we have designed a competition to characterize and rank the performance of these methods when analyzing the dynamic behavior of single molecules. To run this competition, we have implemented a software library to simulate realistic data corresponding to widespread diffusion and interaction models, both in the form of trajectories and videos obtained in typical experimental conditions. The competition will constitute the first assessment of these methods, provide insights into the current limits of the field, foster the development of new approaches, and guide researchers to identify optimal tools for analyzing their experiments.

    @article{munoz2023quantitative,
  title = {Quantitative evaluation of methods to analyze motion changes in single-particle experiments (AnDi Challenge 2)},
  author = {Mu{\~n}oz-Gil, Gorka and Bachimanchi, Harshith and Pineda, Jes{\'u}s [...] and Volpe, Giovanni and Manzo, Carlo},
  journal = {Nature Communications},
  arxiv = {2311.02041},
  paper_image = {andi2_logo.png},
  bibtex_show = {true},
  year = {2025},
  html = {https://doi.org/10.1038/s41467-025-61949-x},
  abbr = {Nat. Commun.},
  selected = {true}
}
  3. Learning minimal representations of many-body physics from snapshots of a quantum simulator
    Møller, Frederik, Fernández-Fernández, Gabriel, Schweigler, Thomas, Schoulepnikoff, Paulin, Schmiedmayer, Jörg and Muñoz-Gil, Gorka
    Phys. Rev. Res. Physical Review Research · 2026
    arXiv link

    Analog quantum simulators provide access to many-body dynamics beyond the reach of classical computation. However, extracting physical insights from experimental data is often hindered by measurement noise, limited observables, and incomplete knowledge of the underlying microscopic model. Here, we develop a machine learning approach based on a variational autoencoder (VAE) to analyze interference measurements of tunnel-coupled one-dimensional Bose gases, which realize the sine-Gordon quantum field theory. Trained in an unsupervised manner, the VAE learns a minimal latent representation that strongly correlates with the equilibrium control parameter of the system. Applied to nonequilibrium protocols, the latent space uncovers signatures of frozen-in solitons following rapid cooling, and reveals anomalous postquench dynamics not captured by conventional correlation-based methods. These results demonstrate that generative models can extract physically interpretable variables directly from noisy and sparse experimental data, providing complementary probes of equilibrium and nonequilibrium physics in quantum simulators. More broadly, our work highlights how machine learning can supplement established field-theoretical techniques, paving the way for scalable, data-driven discovery in quantum many-body systems.

All publications →