AI Insight
Researchers developed PEM-UDE, a machine learning method that discovers mathematical equations governing chaotic systems from noisy experimental data by combining prediction-error methodology with universal differential equations. The technique successfully recovered equations from benchmark chaotic systems even with high noise levels and was applied to neural populations, generating a multi-scale model linking individual neuron parameters to network-level dynamics. The model predicted relationships between connection density, oscillation frequency, and synchrony that showed consistency with intracranial recordings from rat and human cortices.
Why it matters
This method could enable scientists to extract interpretable mathematical models from complex biological and physical systems where data is limited and noisy. For neuroscience specifically, it provides a framework to connect microscopic neuron properties to macroscopic brain activity patterns, potentially improving our understanding of neural circuit function and dysfunction.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
-cross
Abstract: Extracting interpretable mathematical models from complex dynamical systems is difficult, especially for chaotic dynamics observed with noisy experimental data. We present PEM-UDE, a method that combines prediction-error methodology with universal differential equations to discover governing equations from limited, noise-corrupted observations. Prediction-error feedback smooths the chaotic optimization problem; for noise-free data generated within the model class, it preserves the data-consistent zero-loss set, whereas noise and model misspecification introduce a gain-dependent stability-bias trade-off. We test the method on two benchmark chaotic systems, the Rossler attractor and a real electrical circuit, and recover the correct functional forms even when one observed dimension contains noise of five times the signal magnitude. The method also accepts prior knowledge of the system as an initial functional form, which we use to learn neural circuit equations that account for sparse connectivity, a feature missing from conventional neural mass models. Applied to a population of Izhikevich neurons, PEM-UDE yields a multi-scale neural mass model that ties single-neuron parameters to macroscopic network dynamics and predicts a relationship between connection density, dominant oscillation frequency, and synchrony. We test these predictions against three intracranial recording datasets from rat and human cortices. For the neuroscience application, the learned equations are a reduced-order closure for a specified simulated Izhikevich network family; the experimental recordings provide an indirect consistency check of predicted frequency and synchrony trends, not a direct fit of the equations to recordings.
Source: Scientific Machine Learning of Chaotic Systems Learns Reduced-Order Equations for Neural Populations