Non-Stationary Dynamic Mode Decomposition

Many physical processes display complex high-dimensional time-varying behavior, from global weather patterns to brain activity. An outstanding challenge is to express high dimensional data in terms of a dynamical model that reveals their spatiotemporal structure. Dynamic Mode Decomposition is a means to achieve this goal, allowing the identification of key spatiotemporal modes through the diagonalization of a finite dimensional approximation of the Koopman operator. However, DMD methods apply best to time-translationally invariant or stationary data, while in many typical cases, dynamics vary across time and conditions. To capture this temporal evolution, we developed a method, Non-Stationary Dynamic Mode Decomposition (NS-DMD), that generalizes DMD by fitting global modulations of drifting spatiotemporal modes. This method accurately predicts the temporal evolution of modes in simulations and recovers previously known results from simpler methods. To demonstrate its properties, the method is applied to multi-channel recordings from an awake behaving non-human primate performing a cognitive task.

[1]  Bijan Pesaran,et al.  Multi-block RNN Autoencoders Enable Broadband ECoG Signal Reconstruction , 2022, bioRxiv.

[2]  Zhuoyuan Song,et al.  Koopman Operator Theory for Nonlinear Dynamic Modeling using Dynamic Mode Decomposition , 2021, ArXiv.

[3]  J. Kutz,et al.  Bagging, optimized dynamic mode decomposition for robust, stable forecasting with spatial and temporal uncertainty quantification , 2021, Philosophical Transactions of the Royal Society A.

[4]  Emily B. Fox,et al.  Granger Causality: A Review and Recent Advances , 2021, Annual review of statistics and its application.

[5]  S. Brunton,et al.  Modern Koopman Theory for Dynamical Systems , 2021, SIAM Rev..

[6]  K. Soman,et al.  Temperature forecasting using Dynamic Mode Decomposition , 2020, 2020 3rd International Conference on Intelligent Sustainable Systems (ICISS).

[7]  Emery N. Brown,et al.  PLSO: A generative framework for decomposing nonstationary time-series into piecewise stationary oscillatory components , 2020, UAI.

[8]  Scott W. Linderman,et al.  Recurrent Switching Dynamical Systems Models for Multiple Interacting Neural Populations , 2020, bioRxiv.

[9]  E. Koechlin,et al.  Neural mechanisms resolving exploitation-exploration dilemmas in the medial prefrontal cortex , 2020, Science.

[10]  Ivan Smetannikov,et al.  Feature Selection Algorithms as One of the Python Data Analytical Tools , 2020, Future Internet.

[11]  Dimitrios Giannakis,et al.  Data-driven Koopman operator approach for computational neuroscience , 2019, Annals of Mathematics and Artificial Intelligence.

[12]  Selin Aviyente,et al.  Time-Frequency Based Phase-Amplitude Coupling Measure For Neuronal Oscillations , 2019, Scientific Reports.

[13]  J. Gore,et al.  On the Relationship between MRI and Local Field Potential Measurements of Spatial and Temporal Variations in Functional Connectivity , 2019, Scientific Reports.

[14]  Rajesh P. N. Rao,et al.  Time-varying Autoregression with Low Rank Tensors , 2019, SIAM J. Appl. Dyn. Syst..

[15]  R. Kerswell,et al.  Koopman mode expansions between simple invariant solutions , 2018, Journal of Fluid Mechanics.

[16]  Igor Mezic,et al.  Koopman Operator Family Spectrum for Nonautonomous Systems , 2018, SIAM J. Appl. Dyn. Syst..

[17]  Greg J. Stephens,et al.  Adaptive, locally linear models of complex dynamics , 2018, Proceedings of the National Academy of Sciences.

[18]  Demba Ba,et al.  State-space multitaper time-frequency analysis , 2017, Proceedings of the National Academy of Sciences.

[19]  L. Waldorp,et al.  A Tutorial on Estimating Time-Varying Vector Autoregressive Models , 2017, Multivariate behavioral research.

[20]  Chethan Pandarinath,et al.  Inferring single-trial neural population dynamics using sequential auto-encoders , 2017, Nature Methods.

[21]  J. Nathan Kutz,et al.  Variable Projection Methods for an Optimized Dynamic Mode Decomposition , 2017, SIAM J. Appl. Dyn. Syst..

[22]  Bingni W. Brunton,et al.  Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems , 2016 .

[23]  Chethan Pandarinath,et al.  LFADS - Latent Factor Analysis via Dynamical Systems , 2016, ArXiv.

[24]  S. Treue,et al.  Attention Decreases Phase-Amplitude Coupling, Enhancing Stimulus Discriminability in Cortical Area MT , 2015, Front. Neural Circuits.

[25]  P. Fries.  Rhythms for Cognition: Communication through Coherence , 2015, Neuron.

[26]  Steven L. Brunton,et al.  Multiresolution Dynamic Mode Decomposition , 2015, SIAM J. Appl. Dyn. Syst..

[27]  A. Seth,et al.  Granger Causality Analysis in Neuroscience and Neuroimaging , 2015, The Journal of Neuroscience.

[28]  Clarence W. Rowley,et al.  Characterizing and correcting for the effect of sensor noise in the dynamic mode decomposition , 2014, Experiments in Fluids.

[29]  Steven L. Brunton,et al.  Dynamic Mode Decomposition with Control , 2014, SIAM J. Appl. Dyn. Syst..

[30]  Bingni W. Brunton,et al.  Extracting spatial–temporal coherent patterns in large-scale neural recordings using dynamic mode decomposition , 2014, Journal of Neuroscience Methods.

[31]  Yvonne Freer,et al.  Autoregressive Hidden Markov Models for the Early Detection of Neonatal Sepsis , 2014, IEEE Journal of Biomedical and Health Informatics.

[32]  Biyu J. He.  Scale-free brain activity: past, present, and future , 2014, Trends in Cognitive Sciences.

[33]  Jonathan H. Tu,et al.  On dynamic mode decomposition: Theory and applications , 2013, 1312.0041.

[34]  Michael J. Jutras,et al.  Oscillatory activity in the monkey hippocampus during visual exploration and memory formation , 2013, Proceedings of the National Academy of Sciences.

[35]  Tao Zhang,et al.  Reduction in LFP cross-frequency coupling between theta and gamma rhythms associated with impaired STP and LTP in a rat model of brain ischemia , 2013, Front. Comput. Neurosci..

[36]  L. Avery,et al.  The Geometry of Locomotive Behavioral States in C. elegans , 2013, PloS one.

[37]  Verónica Bolón-Canedo,et al.  A review of feature selection methods on synthetic data , 2013, Knowledge and Information Systems.

[38]  Adrián Ponce-Alvarez,et al.  Context-related frequency modulations of macaque motor cortical LFP beta oscillations. , 2012, Cerebral cortex.

[39]  C. Koch,et al.  The origin of extracellular fields and currents — EEG, ECoG, LFP and spikes , 2012, Nature Reviews Neuroscience.

[40]  I. Mezić,et al.  Spectral analysis of nonlinear flows , 2009, Journal of Fluid Mechanics.

[41]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[42]  Michael I. Jordan,et al.  Nonparametric Bayesian Learning of Switching Linear Dynamical Systems , 2008, NIPS.

[43]  P. Schmid,et al.  Dynamic mode decomposition of numerical and experimental data , 2008, Journal of Fluid Mechanics.

[44]  Chih-Jen Lin,et al.  Projected Gradient Methods for Nonnegative Matrix Factorization , 2007, Neural Computation.

[45]  B. R. Noack,et al.  A hierarchy of low-dimensional models for the transient and post-transient cylinder wake , 2003, Journal of Fluid Mechanics.

[46]  W. Freeman,et al.  Aperiodic phase resetting in scalp EEG of beta-gamma oscillations by state transitions at alpha-theta rates , 2003 .

[47]  Thomas M. Smith,et al.  An Improved In Situ and Satellite SST Analysis for Climate , 2002 .

[48]  小羽 俊士,et al.  Wisconsin card sorting test , 2013 .

[49]  H S Seung,et al.  How the brain keeps the eyes still. , 1996, Proceedings of the National Academy of Sciences of the United States of America.

[50]  K. Zhang,et al.  Representation of spatial orientation by the intrinsic dynamics of the head-direction cell ensemble: a theory , 1996, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[51]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[52]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[53]  Gene H. Golub,et al.  The differentiation of pseudo-inverses and non-linear least squares problems whose variables separate , 1972, Milestones in Matrix Computation.

[54]  David Durand,et al.  The Distribution of Length and Components of the Sum of $n$ Random Unit Vectors , 1955 .

[55]  B. O. Koopman,et al.  Dynamical Systems of Continuous Spectra. , 1932, Proceedings of the National Academy of Sciences of the United States of America.

[56]  Scott W. Linderman,et al.  Bayesian Learning and Inference in Recurrent Switching Linear Dynamical Systems , 2017, AISTATS.

[57]  Clarence W. Rowley,et al.  Extending Data-Driven Koopman Analysis to Actuated Systems , 2016 .

[58]  Stephen E. Levinson,et al.  Autoregressive Hidden Markov Model and the Speech Signal , 2015, Complex Adaptive Systems.

[59]  A. Provenzale.  Climate as a Complex Dynamical System , 2014 .

[60]  G. Buzsáki.  Rhythms of the brain , 2006 .

[61]  Ning Qian,et al.  On the momentum term in gradient descent learning algorithms , 1999, Neural Networks.

[62]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .