Artificial intelligence as a surrogate brain: Bridging neural dynamical models and data
🤖 AI Summary
This study addresses the critical challenge of whole-brain dynamical forecasting by proposing a unified computational framework that treats AI models as “surrogate brains.” Methodologically, it formulates deep learning models as learnable dynamical system agents, integrating forward modeling, inverse problem solving, and multi-scale neurodynamic constraints to decode latent brain states from large-scale neural signals and accurately forecast nonlinear whole-brain evolution. The key contribution is the first development of a closed-loop computational platform that is simultaneously interpretable, intervenable, and simulatable—enabling virtual perturbation analysis and model-guided neuromodulation. Results demonstrate a principled bridge between theoretical neuroscience and translational neuroengineering, with significant implications for personalized intervention in neurological disorders and next-generation brain–computer interfaces.
Application Category
📝 Abstract
Recent breakthroughs in artificial intelligence (AI) are reshaping the way we construct computational counterparts of the brain, giving rise to a new class of ``surrogate brains''. In contrast to conventional hypothesis-driven biophysical models, the AI-based surrogate brain encompasses a broad spectrum of data-driven approaches to solve the inverse problem, with the primary objective of accurately predicting future whole-brain dynamics with historical data. Here, we introduce a unified framework of constructing an AI-based surrogate brain that integrates forward modeling, inverse problem solving, and model evaluation. Leveraging the expressive power of AI models and large-scale brain data, surrogate brains open a new window for decoding neural systems and forecasting complex dynamics with high dimensionality, nonlinearity, and adaptability. We highlight that the learned surrogate brain serves as a simulation platform for dynamical systems analysis, virtual perturbation, and model-guided neurostimulation. We envision that the AI-based surrogate brain will provide a functional bridge between theoretical neuroscience and translational neuroengineering.
Problem
Research questions and friction points this paper is trying to address.
Predicting future whole-brain dynamics using historical data
Bridging theoretical neuroscience with translational neuroengineering applications
Solving inverse problems through data-driven AI approaches
Innovation
Methods, ideas, or system contributions that make the work stand out.
AI-based surrogate brain for neural dynamics prediction
Unified framework integrating forward and inverse modeling
Simulation platform for virtual perturbation and neurostimulation
🔎 Similar Papers
2024-01-30arXiv.orgCitations: 4
