Ábel Ságodi

PhD Student in CATNIPLAB at Champalimaud Research, supervised by Memming Park.

abel_sagodi.jpg

Hi, I’m Ábel Ságodi. I am a PhD student at the Champalimaud Centre for the Unknown in Lisbon, working at the intersection of theoretical neuroscience, machine learning, and control theory. With a background in Mathematics, Physics, and Computational Science (MSc & BSc from the University of Amsterdam), I approach the brain as a complex dynamical system. I am fascinated by how biological circuits, despite being noisy and biological, can perform robust computation. I draw inspiration from the philosophy of science to develop an interpretable language for this computation, finding the right mathematical abstractions to understand how the brain actually works.

news

Sep 28, 2026 Our paper Universal Approximation Theorems for Dynamical Systems with Infinite-Time Horizon Guarantees got accepted at NeurIPS 2026! This is joint work with Memming Park.
Jul 15, 2026 Our paper Data-Enabled Local Reach-Avoid Control Near Saddle Equilibria with a Neuronal Interpretation got accepted for the 65th IEEE Conference on Decision and Control (CDC)! This is joint work with Joshua L. Pughe-Sanford, Xuehao Ding and Dmitri B. Chklovskii.
May 10, 2026 Two posters were accepted for Dynamics Days Europe! The first one is Universal Approximation Theorems for Dynamical Systems with Infinite-Time Horizon Guarantees in collaboration with Memming Park, while the second is Piecewise Projective Maps for Analytical Solutions of Spiking Neural Network Dynamics with Ildefonso Ferreira Pica and Bill Podlaski.

selected publications

  1. NeurIPS
    uap_landscape.png
    Universal Approximation Theorems for Dynamical Systems with Infinite-Time Horizon Guarantees
    Ábel Ságodi and I. M. Park
    In Fortieth Annual Conference on Neural Information Processing Systems (to appear), 2026
    arXiv HTML
  2. arXiv
    topological_conjugacy_pareto.png
    Dynamical Archetype Analysis: Autonomous Computation
    Ábel Ságodi and I. M. Park
    arXiv preprint arXiv:2507.05505, 2025
    arXiv HTML
  3. NeurIPS
    FenichelThm.png
    Back to the Continuous Attractor
    Ábel Ságodi, G. Martin-Sanchez, P. A. Sokól, and I. M. Park
    In Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024
    HTML