Identifying Latent Neural Dynamics with Recognition-Parametrised Gaussian Process Dynamical Systems
Organizations: Gatsby Computational Neuroscience Unit University College London London, UK
Abstract
Neural circuits are densely interconnected systems, and many neural computations are thought to be mediated by the dynamical evolution of activity within these recurrent networks, a hypothesis often referred to as "computation through dynamics". Under this view, understanding computation within real neural populations depends on building models of the dynamical rules governing the temporal evolution of recorded high-dimensional neural activity. Most commonly, such models depend on explicit generative parameterisations of (i) the (nonlinear) intrinsic dynamical flow field and stochasticity, (ii) the (nonlinear) mapping from dynamical state to neuronal activity, and (iii) the variability of individual neural responses given that dynamical state. Misspecification of any of these components can bias estimates of dynamics and latent trajectories. In this paper, we circumvent generation-related issues by introducing the Recognition-Parametrised Gaussian Process dynamics (RP-GPdyn) model. RP-GPdyn models the dynamical flow field using a nonparametric Gaussian Process-defined transition function and models the relationship of dynamics to neural activity implicitly using the Recognition-Parametrised Model (RPM) framework. We show that this generation-free approach uncovers meaningful, behaviourally relevant latent variables and dynamics from both synthetic and experimental datasets.
Figures & tables
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
| Symbol | Description |
|---|---|
| Number of observed sequences or trials. | |
| Number of time steps per sequence. | |
| Observation dimension, e.g., number of recorded neurons or neural features. | |
| Latent state dimension. | |
| Number of inducing points used in the sparse GP approximation. | |
| Number of fixed points explicitly parameterizing the GP dynamics prior. |
| Model | Inducing | Optimisation schedule | GPU | Runtime / iteration (s) |
|---|---|---|---|---|
| RP-GPdyn | 30 | 10k grad. updates, batch 32 | T1000 (Turing) | |
| SING | 81 | 50 vEM; 30 E + 50 M, full batch | RTX 4500 (Ada) |
Latent Dynamics Modeling
Latent dynamics modeling learns compact hidden-state representations and transition rules to predict, explain, and control evolving systems.
See all 274 papers