Linear dynamical neural population models through nonlinear embeddings
This work addresses the challenge of accurately modeling neural population dynamics for neuroscientists, offering improved interpretability and performance over existing methods, though it is incremental in extending linear models with nonlinear embeddings.
The authors tackled the problem of modeling neural activity by proposing fLDS, a nonlinear generative model that captures richer neural variability than linear models while retaining an interpretable low-dimensional latent space, and showed it captures a much larger proportion of neural variability with superior predictive performance on two datasets.
A body of recent work in modeling neural activity focuses on recovering low-dimensional latent features that capture the statistical structure of large-scale neural populations. Most such approaches have focused on linear generative models, where inference is computationally tractable. Here, we propose fLDS, a general class of nonlinear generative models that permits the firing rate of each neuron to vary as an arbitrary smooth function of a latent, linear dynamical state. This extra flexibility allows the model to capture a richer set of neural variability than a purely linear model, but retains an easily visualizable low-dimensional latent space. To fit this class of non-conjugate models we propose a variational inference scheme, along with a novel approximate posterior capable of capturing rich temporal correlations across time. We show that our techniques permit inference in a wide class of generative models.We also show in application to two neural datasets that, compared to state-of-the-art neural population models, fLDS captures a much larger proportion of neural variability with a small number of latent dimensions, providing superior predictive performance and interpretability.