Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems

By Scott W. Linderman, Matthew J. Johnson, Andrew C. Miller, Ryan P. Adams, David M. Blei, Liam Paninski. 2017

309 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning (2017, background): “[20] introduces recurrent switching linear dynamical systems, that combine deep learning techniques and switching Kalman filters [22] to model low-dimensional time series.”
Unsupervised Discovery of Demixed, Low-Dimensional Neural Dynamics across Multiple Timescales through Tensor Component Analysis (2017, methodology): “In particular, single trial neural activity has been modeled with linear 517 dynamics [81–84], switched linear dynamics [85, 86], linear dynamics with nonlinear observations [17], and nonlinear 518 dynamics [18, 87].”
Inferring single-trial neural population dynamics using sequential auto-encoders (2017, background): “…(Yu et al. 2009; Petreska et al. 2011; Macke et al. 2011; Kao et al. 2015; Aghagolzadeh and Truccolo 2014; Zhao and Park 2017; Y. Gao et al. 2016; Linderman et al. 2017), but its primary innovations are the use of nonlinear RNNs for the dynamics generator, the ability to infer inputs describing…”
Computation Through Neural Population Dynamics (2020, background): “Through these linear approximations we understand Rθ , and through it f. Models that switch between multiple LDSs (SLDS) have a rich history and also may serve as stand-alone models of f (e.g., Linderman et al. 2017, Petreska et al. 2011).”
Tree-Structured Recurrent Switching Linear Dynamical Systems for Multi-Scale Modeling (2018, methodology): “Linderman et al. (2017) shows how to perform message-passing using the Pólya-gamma augmented recurrence potentials ψ ( x t , z t , w n,t ) .”