Gorka Muñoz-Gil

I'm a Marie Skłodowska-Curie postdoctoral fellow in the Quantum Information and Computation group at the University of Innsbruck (UIBK), Austria.

I got my PhD in the Quantum Optics theory group at ICFO ( November 2020! ). Before that I got a MSc in Photonics at the Universitat Politècnica de Catalunya and BSc in Physics at the Universitat Autònoma de Barcelona .

A longer description of what I have been doing during these years can be found in my CV

Main areas of research

In recent years, I have been captivated by the potential Machine Learning (ML) techniques offer to the study of Physics, and in particular in the design of interpretable solutions that can assist researchers to gain further insight from their experiments!

Anomalous diffusion: from life to machines

Anomalous diffusion was the focus of my PhD Thesis. Motivated by collaborations with Prof. Maria García-Parajo (ICFO) and Prof. Carlo Manzo (UVic), I developed theoretical models to better understand micro- and nano-particle motion in cells. Realizing the potential of ML to bridge experimental observations with theory, we pioneered its use in anomalous diffusion and organized the AnDI Challenge, a scientific competition that has already celebrated two editions.

Since then, I have advanced anomalous diffusion characterization by proposing the STEP method—which probes diffusion at the single-step level—and showing that unsupervised learning can rediscover the fundamental components of our diffusion theories (link). We have also validated these approaches experimentally, gaining new insights into the process of particle condensation (link). Lately, we have studied how artifical learning agents can be used to find diffusion-based optimal foraging strategies (link).

ML in Quantum Physics

I also apply Machine Learning to the quantum realm, starting with the deep connection between Boltzmann machines and spin models, which led to the (RAPID) architecture. We then employed reinforcement learning for the certification of quantum systems (link).

More recently, we introduced powerful denoising diffusion models for quantum circuit synthesis, which earned a cover feature in Nature Machine Intelligence. This approach opens exciting new avenues in quantum computing that I look forward to exploring further.

Science beyond science

Science doesn’t end with publications. I love engaging in diverse projects that extend my research beyond the lab. For instance, I’m part of the Night up project, studying light pollution through citizen science. I also collaborated on a quantum-inspired music piece, performed by Reiko Yamada at the Sonar Festival 2021. In addition, I frequently participate in outreach initiatives for students and the general public—see the Teaching tab for more details!

News

Jun 2025	New pre-print: Interpretable representation learning of quantum data enabled by probabilistic variational autoencoders! In this new work we show how to adapt VAEs, the cornerstone architecture for interpretable AI, to properly deal with quantum data.
Jun 2025	New pre-print: Synthesis of discrete-continuous quantum circuits with multimodal diffusion models! We’re back with quantum circuit synthesis using diffusion models, now supporting parameterized gates! We show how to generate quantum circuits for arbitrary unitaries, but also use those circuits to perform gagdegt extraction to learn new insights!.
Mar 2025	I won the Postdoc Poster Prize award at the AI4Quantum conference for our poster on generating quantum circuits with diffusion models!

Recent selected publications

Proc. Natl. Acad. Sci.
Stochastic particle unbinding modulates growth dynamics and size of transcription factor condensates in living cells

Muñoz-Gil, Gorka, Romero, Catalina, Mateos, Nicolas, Llobet Cucalon, Lara Isabel, Filion, Guillaume, Beato, Miguel, Lewenstein, Maciej, Garcı́ia-Parajo, Marı́ia, and Torreno-Pina, Juan

Proceedings of the National Academy of Sciences 2022

Abstract Bib LINK

Living cells organize internal compartments by forming molecular condensates that operate as versatile biochemical “hubs.” Their occurrence is particularly relevant in the nucleus where they regulate, amongst others, gene transcription. However, the biophysics of transcription factor (TF) condensation remains highly unexplored. Through single-molecule experiments in living cells, theory, and simulations, we assessed the diffusion, growth dynamics, and sizes of TF condensates of the nuclear progesterone receptor (PR). Interestingly, PR condensates obey classical growth dynamics at shorter times but deviate at longer times, reaching finite sizes at steady-state. We demonstrate that condensate growth dynamics and nanoscale-size arrested growth is regulated by molecular escaping from condensates, providing an exquisite control of condensate size in nonequilibrium systems such as living cells.
@article{munoz2020phase, title = {Stochastic particle unbinding modulates growth dynamics and size of transcription factor condensates in living cells}, author = {Mu{\~n}oz-Gil, Gorka and Romero, Catalina and Mateos, Nicolas and de Llobet Cucalon, Lara Isabel and Filion, Guillaume and Beato, Miguel and Lewenstein, Maciej and Garc{\'\i}ia-Parajo, Mar{\'\i}ia and Torreno-Pina, Juan}, bibtex_show = {true}, paper_image = {stochastic_unbinding.png}, journal = {Proceedings of the National Academy of Sciences}, abbr = {Proc. Natl. Acad. Sci.}, volume = {119}, number = {31}, pages = {e2200667119}, year = {2022}, publisher = {National Acad. Sciences}, html = {https://doi.org/10.1073/pnas.2200667119}, selected = true }
i.p.a. Nat. Commun.
Quantitative evaluation of methods to analyze motion changes in single-particle experiments (AnDi Challenge 2)

Muñoz-Gil, Gorka, Bachimanchi, Harshith, Pineda, Jesús, Midtvedt, Benjamin, Lewenstein, Maciej, Metzler, Ralf, Krapf, Diego, Volpe, Giovanni, and Manzo, Carlo

In principle accepted in Nature Communications 2023

Abstract arXiv Bib

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 Midtvedt, Benjamin and Lewenstein, Maciej and Metzler, Ralf and Krapf, Diego and Volpe, Giovanni and Manzo, Carlo}, journal = {In principle accepted in Nature Communications}, arxiv = {2311.02041}, paper_image = {andi2_logo.png}, bibtex_show = {true}, year = {2023}, abbr = {i.p.a. Nat. Commun.}, selected = {true} }
Nat. Mach. Intell.
Quantum circuit synthesis with diffusion models

Fürrutter, Florian, Muñoz-Gil, Gorka, and Briegel, Hans J

Nature Machine Intelligence 2024

Abstract arXiv Bib 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} }