Abstract:The computational properties of neural systems are often thought to be implemented in terms of their network dynamics. Hence, recovering the system dynamics from experimentally observed neuronal time series, like multiple single-unit (MSU) recordings or neuroimaging data, is an important step toward understanding its computations. Ideally, one would not only seek a state space representation of the dynamics, but would wish to have access to its governing equations for in-depth analysis. Recurrent neural networks (RNNs) are a computationally powerful and dynamically universal formal framework which has been extensively studied from both the computational and the dynamical systems perspective. Here we develop a semi-analytical maximum-likelihood estimation scheme for piecewise-linear RNNs (PLRNNs) within the statistical framework of state space models, which accounts for noise in both the underlying latent dynamics and the observation process. The Expectation-Maximization algorithm is used to infer the latent state distribution, through a global Laplace approximation, and the PLRNN parameters iteratively. After validating the procedure on toy examples, the approach is applied to MSU recordings from the rodent anterior cingulate cortex obtained during performance of a classical working memory task, delayed alternation. A model with 5 states turned out to be sufficient to capture the essential computational dynamics underlying task performance, including stimulus-selective delay activity. The estimated models were rarely multi-stable, but rather were tuned to exhibit slow dynamics in the vicinity of a bifurcation point. In summary, the present work advances a semi-analytical (thus reasonably fast) maximum-likelihood estimation framework for PLRNNs that may enable to recover the relevant dynamics underlying observed neuronal time series, and directly link them to computational properties.

Meta-Dynamical State Space Models for Integrative Neural Data Analysis

Towards a theory of learning dynamics in deep state space models

Integrating Multimodal Data for Joint Generative Modeling of Complex Dynamics

Avoiding Catastrophe: Active Dendrites Enable Multi-Task Learning in Dynamic Environments

Learning low-dimensional dynamics from whole-brain data improves task capture

Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics

Decomposed Linear Dynamical Systems (dLDS) for learning the latent components of neural dynamics

Neural State-Space Modeling with Latent Causal-Effect Disentanglement

Meta-State-Space Learning: An Identification Approach for Stochastic Dynamical Systems

Physics-Informed Neural State Space Models via Learning and Evolution

Deep multi-modal learning for joint linear representation of nonlinear dynamical systems

Belief dynamics extraction

A State Space Approach for Piecewise-Linear Recurrent Neural Networks for Reconstructing Nonlinear Dynamics from Neural Measurements

Generative Modeling of Neural Dynamics via Latent Stochastic Differential Equations

State-space model with deep learning for functional dynamics estimation in resting-state fMRI

A Distributed Neural Hybrid System Learning Framework in Modeling Complex Dynamical Systems

Extracting low-dimensional dynamics from multiple large-scale neural population recordings by learning to predict correlations

Real-Time Machine Learning Strategies for a New Kind of Neuroscience Experiments

Learning Latent Dynamics via Invariant Decomposition and (Spatio-)Temporal Transformers

Meta Neural Coordination