LFADS - Latent Factor Analysis via Dynamical Systems

2016/08/22 by David Sussillo, Rafal Jozefowicz, Sussillo, David +5 · 5 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · #FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neurons and Cognition (q-bio.NC) #cs.LG #q-bio.NC #stat.ML

paper · pdf · doi:10.48550/arxiv.1608.06315

16 pages, 11 figures

arxiv created 2016/08/22 · arxiv updated 2016/08/24

Abstract

Neuroscience is experiencing a data revolution in which many hundreds or thousands of neurons are recorded simultaneously. Currently, there is little consensus on how such data should be analyzed. Here we introduce LFADS (Latent Factor Analysis via Dynamical Systems), a method to infer latent dynamics from simultaneously recorded, single-trial, high-dimensional neural spiking data. LFADS is a sequential model based on a variational auto-encoder. By making a dynamical systems hypothesis regarding the generation of the observed data, LFADS reduces observed spiking to a set of low-dimensional temporal factors, per-trial initial conditions, and inferred inputs. We compare LFADS to existing methods on synthetic data and show that it significantly out-performs them in inferring neural firing rates and latent dynamics.

Cited by

• Self-Supervised Dynamical System Representations for Physiological Time-Series
• Energy-based Autoregressive Generation for Neural Population Dynamics
• A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks
• Inpainting the Neural Picture: Inferring Unrecorded Brain Area Dynamics from Multi-Animal Datasets
• Langevin Flows for Modeling Neural Latent Dynamics

Related

• lfads-torch: A modular and extensible implementation of latent factor analysis via dynamical systems
• Accurate identification of communication between multiple interacting neural populations
• Functional Connectivity: Continuous-Time Latent Factor Models for Neural Spike Trains
• Linear dynamical neural population models through nonlinear embeddings
• Multi-modal Gaussian Process Variational Autoencoders for Neural and Behavioral Data