The Budapest Computational Neuroscience Forum is a series of informal monthly meetings of Budapest-based computational neuroscientists and computational cognitive scientists with the aim of facilitating discussion and cooperation among researchers working in different institutes and giving an opportunity to students to present their work and get to know the comunity. Originally started in 2007, restarted in 2017 and then again in 2023 the Forum is now regularly hosted by Central European University, and followed by a social event, both open to anyone interested.
Events of the Forum are advertised on a mailing list. If you wish to be on this list or have any inquiries about the series, contact Mihály Bányai.
Upcoming meeting:
Time: November 11, Tuesday, 17:00
Location: CEU, 1051 Bp. Nádor u. 15, Room 105
Presenter: Zsigmond Benkő, Wigner Institute
Title: Hidden Common Driver Reconstruction and the Anisotropic Self-Organizing Map
Abstract:
Hidden common signals drive many coupled systems yet remain unobserved. Reconstructing these latent drivers from observed time series is central to understanding coordination, building invariant representations, and denoising dynamical data. In this talk, I will approach the problem of hidden driver reconstruction from a dynamical-systems perspective and outline a progression of methods that aim to uncover shared structure across systems—from Slow Feature Analysis and Canonical Correlation Analysis to their deep extensions, shared dynamics reconstruction, and the Mapper–Coach (MaCo) framework. Building on this context, I will introduce the Self-Organizing Map (SOM) and present our Anisotropic SOM (ASOM) variant [1], which adapts the classical algorithm to learn the geometry of coupled attractors and separate self-dynamics from shared dynamics. While the focus will be on the mathematical and algorithmic foundations, the results may also hint at broader connections to how coordination or shared patterns emerge in neural and cognitive systems.
[1] Benkő, Zsigmond, Marcell Stippinger, Attila Bencze, Fülöp Bazsó, András Telcs, and Zoltán Somogyvári. “Inference of Hidden Common Driver Dynamics by Anisotropic Self-Organizing Neural Networks.” Neural Networks 194 (February 2026): 108113. https://doi.org/10.1016/j.neunet.2025.108113.
Earlier meetings of the Forum:
June 17, 2025. Keith T Murray (Wigner Institute): V1 signatures of generative computations in a discrimination task
November 5, 2024. József Konczer: Statistical Games, Playful approach to statistics. Recording here.
May 22, 2024. Nikola Milićević (Pennsylvania State University): Sensory systems and combinatorial neural codes. Recording here.
January 31, 2024. Zoltán Somogyvári (Wigner Institute): Seeing beyond the spikes: reconstructing the complete spatiotemporal membrane potential distribution from paired intra- and extracellular recordings
December 12, 2023. Máté Lengyel (Cambridge/CEU): Optimal information loading into working memory explains dynamic coding in prefrontal cortex. Recording here.
November 15, 2023. András Ecker (EPFL): Long-term plasticity induces sparse and specific synaptic changes in a biophysically detailed cortical model
October 25, 2023. Ferenc Csikor (Wigner Institute): Top-down perceptual inference shaping the activity of early visual cortex
October 4, 2023. Gábor Lengyel (University of Rochester): A General Method for Testing Bayesian Models Using Neural Data
September 1, 2023. Emmanuel Procyk (INSERM Lyon): Prefrontal neuronal dynamics, timescales, and behavioural flexibility
May 31, 2023. Atilla B. Kelemen (KOKI): Geometry of remapping in the hippocampus reflects task structure
April 26, 2023. Anna Székely (Wigner Institute): Identifying transfer learning in the reshaping of inductive biases
March 29, 2023. Merse Előd Gáspár (CEU): Discovering the internal predictive model of infants based on eye movements