Artificial intelligence as a surrogate brain: Bridging neural dynamical models and data

2025/10/11 by Yinuo Zhang, Zhang, Yinuo, Zhichao Liang +14
Neuroscience · Physics and Astronomy · #Functional Brain Connectivity Studies #Neural dynamics and brain function #Model Reduction and Neural Networks

paper · pdf · doi:10.48550/arxiv.2510.10308

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.

Citations

• Dynamic-Attention-based EEG State Transition Modeling for Emotion Recognition
• A scalable generative model for dynamical system reconstruction from neuroimaging data
• Understanding Auditory Evoked Brain Signal via Physics-informed Embedding Network with Multi-Task Transformer
• Reverse engineering the brain input: Network control theory to identify cognitive task-related control nodes
• A Direct Sampling Method and Its Integration with Deep Learning for Inverse Scattering Problems with Phaseless Data
• BrainMass: Advancing Brain Network Analysis for Diagnosis with Large-scale Self-Supervised Learning
• BrainWave: A Brain Signal Foundation Model for Clinical Applications
• CLIP-MUSED: CLIP-Guided Multi-Subject Visual Neural Information Semantic Decoding
• Interacting Particle Systems on Networks: joint inference of the network and the interaction kernel
• Advancing EEG/MEG Source Imaging with Geometric-Informed Basis Functions
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces
• Neuro-GPT: Towards A Foundation Model for EEG
• SpikingJelly: An open-source machine learning infrastructure platform for spike-based intelligence
• Sparse Deep Learning for Time Series Data: Theory and Applications
• Persistent learning signals and working memory without continuous attractors
• Low Tensor Rank Learning of Neural Dynamics
• RMP-Loss: Regularizing Membrane Potential Distribution for Spiking Neural Networks
• Perturbing a Neural Network to Infer Effective Connectivity: Evidence from Synthetic EEG Data
• FinGPT: Democratizing Internet-scale Data for Financial Large Language Models
• Reservoir Computing with Error Correction: Long-term Behaviors of Stochastic Dynamical Systems
• GPT-4 Technical Report
• BrainCLIP: Bridging Brain and Visual-Linguistic Representation Via CLIP for Generic Natural Visual Stimulus Decoding
• Causal Recurrent Variational Autoencoder for Medical Time Series Generation
• Mapping effective connectivity by virtually perturbing a surrogate brain
• Learning Task-Aware Effective Brain Connectivity for fMRI Analysis with Graph Neural Networks
• Explainable fMRI-based Brain Decoding via Spatial Temporal-pyramid Graph Convolutional Network
• A Novel Semi-supervised Meta Learning Method for Subject-transfer Brain-computer Interface
• Algorithms to estimate Shapley value feature attributions
• Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systems
• On the effectiveness of persistent homology
• Training language models to follow instructions with human feedback
• Generalized Out-of-Distribution Detection: A Survey
• Generalized Out-of-Distribution Detection: A Survey
• On the difficulty of learning chaotic dynamics with RNNs
• Optimal input representation in neural systems at the edge of chaos
• Online Learning Koopman operator for closed-loop electrical neurostimulation in epilepsy
• Learning Transferable Visual Models From Natural Language Supervision
• HNPE: Leveraging Global Parameters for Neural Posterior Estimation
• Interpretable multimodal fusion networks reveal mechanisms of brain cognition
• WeatherBench: A benchmark dataset for data-driven weather forecasting
• Learning Stable Deep Dynamics Models
• Hamiltonian Neural Networks
• Learning to Adaptively Scale Recurrent Neural Networks
• Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Neural Ordinary Differential Equations
• Disentangling by Factorising
• Dynamic models of large-scale brain activity
• Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
• A general framework for stochastic traveling waves and patterns, with application to neural field equations
• Multi-Resolution Dynamic Mode Decomposition
• NeuCube: A spiking neural network architecture for mapping, learning and understanding of spatio-temporal brain data
• Uncertainty quantification and weak approximation of an elliptic inverse problem
• Electroencephalogram and visual evoked potential generation in a mathematical model of coupled cortical columns
• A neurophysiologically-based mathematical model of flash visual evoked potentials

Related

• A brain basis of dynamical intelligence for AI and computational neuroscience
• A New Strategy for Artificial Intelligence: Training Foundation Models Directly on Human Brain Data
• Biological Blueprints for Next Generation AI Systems
• The brain versus AI: World-model-based versatile circuit computation underlying diverse functions in the neocortex and cerebellum
• NeuroAI and Beyond: Bridging Between Advances in Neuroscience and ArtificialIntelligence