Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAE

Abstract

The ability to record activities from hundreds of neurons simultaneously in the brain has placed an increasing demand for developing appropriate statistical techniques to analyze such data. Recently, deep generative models have been proposed to fit neural population responses. While these methods are flexible and expressive, the downside is that they can be difficult to interpret and identify. To address this problem, we propose a method that integrates key ingredients from latent models and traditional neural encoding models. Our method, pi-VAE, is inspired by recent progress on identifiable variational auto-encoder, which we adapt to make appropriate for neuroscience applications. Specifically, we propose to construct latent variable models of neural activity while simultaneously modeling the relation between the latent and task variables (non-neural variables, e.g. sensory, motor, and other externally observable states). The incorporation of task variables results in models that are not only more constrained, but also show qualitative improvements in interpretability and identifiability. We validate pi-VAE using synthetic data, and apply it to analyze neurophysiological datasets from rat hippocampus and macaque motor cortex. We demonstrate that pi-VAE not only fits the data better, but also provides unexpected novel insights into the structure of the neural codes.

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• NICE: Non-linear Independent Components Estimation
• Density estimation using Real NVP
• Demixed principal component analysis of population activity in higher cortical areas reveals independent representation of task parameters
• Composing graphical models with neural networks for structured representations and fast inference
• Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)
• Fixing a Broken ELBO
• Learning interpretable continuous-time models of latent stochastic dynamical systems
• Identifying through Flows for Recovering Latent Representations

Cited by in corpus (6)

• Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity
• Building population models for large-scale neural recordings: opportunities and pitfalls
• Targeted Neural Dynamical Modeling
• Disentangling Identifiable Features from Noisy Data with Structured Nonlinear ICA
• Mine Your Own vieW: Self-Supervised Learning Through Across-Sample Prediction
• Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity