SrVARM: State Regularized Vector Autoregressive Model for Joint Learning of Hidden State Transitions and State-Dependent Inter-Variable Dependencies from Multi-variate Time Series

Many applications, e.g., healthcare, education, call for effective methods methods for constructing predictive models from high dimensional time series data where the relationship between variables can be complex and vary over time. In such settings, the underlying system undergoes a sequence of unobserved transitions among a finite set of hidden states. Furthermore, the relationships between the observed variables and their temporal dynamics may depend on the hidden state of the system. To further complicate matters, the hidden state sequences underlying the observed data from different individuals may not be aligned relative to a common frame of reference. Against this background, we consider the novel problem of jointly learning the state-dependent inter-variable relationships as well as the pattern of transitions between hidden states from multi-variate time series data. To solve this problem, we introduce the State-Regularized Vector Autoregressive Model (SrVARM) which combines a state-regularized recurrent neural network to learn the dynamics of transitions between discrete hidden states with an augmented autoregressive model which models the inter-variable dependencies in each state using a state-dependent directed acyclic graph (DAG). We propose an efficient algorithm for training SrVARM by leveraging a recently introduced reformulation of the combinatorial problem of optimizing the DAG structure with respect to a scoring function into a continuous optimization problem. We report results of extensive experiments with simulated data as well as a real-world benchmark that show that SrVARM outperforms state-of-the-art baselines in recovering the unobserved state transitions and discovering the state-dependent relationships among variables.

[1]  Le Song,et al.  Time-Varying Dynamic Bayesian Networks , 2009, NIPS.

[2]  Mehrdad Nourani,et al.  A Non-EEG Biosignals Dataset for Assessment and Visualization of Neurological Status , 2016, 2016 IEEE International Workshop on Signal Processing Systems (SiPS).

[3]  Alexander J. Hartemink,et al.  Non-stationary dynamic Bayesian networks , 2008, NIPS.

[4]  Keiji Kanazawa,et al.  A model for reasoning about persistence and causation , 1989 .

[5]  Marloes H. Maathuis,et al.  Structure Learning in Graphical Modeling , 2016, 1606.02359.

[6]  Stuart J. Russell,et al.  Dynamic bayesian networks: representation, inference and learning , 2002 .

[7]  Vasant Honavar,et al.  Functional Autoencoders for Functional Data Representation Learning , 2021, SDM.

[8]  Vasant Honavar,et al.  LMLFM: Longitudinal Multi-Level Factorization Machine , 2020, AAAI.

[9]  Juan Zhou,et al.  Learning effective brain connectivity with dynamic Bayesian networks , 2007, NeuroImage.

[10]  Emily B. Fox,et al.  An Interpretable and Sparse Neural Network Model for Nonlinear Granger Causality Discovery , 2017, 1711.08160.

[11]  Sungjin Ahn,et al.  Sequential Neural Processes , 2019, NeurIPS.

[12]  Mihaela van der Schaar,et al.  Attentive State-Space Modeling of Disease Progression , 2019, NeurIPS.

[13]  Yang Zhou,et al.  Structure Learning of Probabilistic Graphical Models: A Comprehensive Survey , 2011, ArXiv.

[14]  Kun Zhang,et al.  Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models , 2019, ICML.

[15]  Suhang Wang,et al.  Explainable Multivariate Time Series Classification: A Deep Neural Network Which Learns to Attend to Important Variables As Well As Time Intervals , 2021, WSDM.

[16]  Jorge Nocedal,et al.  Algorithm 778: L-BFGS-B: Fortran subroutines for large-scale bound-constrained optimization , 1997, TOMS.

[17]  L. Tran,et al.  Integrated Systems Approach Identifies Genetic Nodes and Networks in Late-Onset Alzheimer’s Disease , 2013, Cell.

[18]  Rangan Gupta,et al.  The impact of oil price on South African GDP growth: A Bayesian Markov Switching-VAR analysis , 2017 .

[19]  T. Li,et al.  Impact of situational variables and demographic attributes in two seasons on purchase behaviour , 2000 .

[20]  T. Koski,et al.  A Review of Bayesian Networks and Structure Learning , 2012 .

[21]  Xianfeng Tang,et al.  Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values , 2019, AAAI.

[22]  Matthias W. Seeger,et al.  Deep State Space Models for Time Series Forecasting , 2018, NeurIPS.

[23]  M. Gerstein,et al.  Genomic analysis of regulatory network dynamics reveals large topological changes , 2004, Nature.

[24]  Ian Maconochie,et al.  The relationship between body temperature, heart rate and respiratory rate in children , 2009, Emergency Medicine Journal.

[25]  E. Fox,et al.  Neural Granger Causality for Nonlinear Time Series , 2018, 1802.05842.

[26]  Fabio Stella,et al.  A survey on Bayesian network structure learning from data , 2019, Progress in Artificial Intelligence.

[27]  P. Spirtes,et al.  Review of Causal Discovery Methods Based on Graphical Models , 2019, Front. Genet..

[28]  Qiang Shen,et al.  Learning Bayesian networks: approaches and issues , 2011, The Knowledge Engineering Review.

[29]  David Heckerman,et al.  A Tutorial on Learning with Bayesian Networks , 1999, Innovations in Bayesian Networks.

[30]  Richard E. Turner,et al.  Neural Adaptive Sequential Monte Carlo , 2015, NIPS.

[31]  Vasant Honavar,et al.  Longitudinal Deep Kernel Gaussian Process Regression , 2020, ArXiv.

[32]  Mo Yu,et al.  DAG-GNN: DAG Structure Learning with Graph Neural Networks , 2019, ICML.

[33]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[34]  Alexander J. Smola,et al.  State Space LSTM Models with Particle MCMC Inference , 2017, ArXiv.

[35]  Vasant Honavar,et al.  Dynamical Gaussian Process Latent Variable Model for Representation Learning from Longitudinal Data , 2020, FODS.

[36]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[37]  Cheng Wang,et al.  State-Regularized Recurrent Neural Networks , 2019, ICML.

[38]  Bryon Aragam,et al.  DYNOTEARS: Structure Learning from Time-Series Data , 2020, AISTATS.

[39]  Marco Grzegorczyk,et al.  Non-stationary continuous dynamic Bayesian networks , 2009, NIPS.

[40]  K. Sachs,et al.  Causal Protein-Signaling Networks Derived from Multiparameter Single-Cell Data , 2005, Science.

[41]  Joel Nothman,et al.  SciPy 1.0-Fundamental Algorithms for Scientific Computing in Python , 2019, ArXiv.

[42]  W. Roth,et al.  Methodological considerations in ambulatory skin conductance monitoring. , 2011, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[43]  Richard E. Neapolitan,et al.  Learning Bayesian networks , 2007, KDD '07.

[44]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[45]  Kash Barker,et al.  A Bayesian network model for resilience-based supplier selection , 2016 .

[46]  Zhitang Chen,et al.  Causal Discovery with Reinforcement Learning , 2019, ICLR.

[47]  Pradeep Ravikumar,et al.  DAGs with NO TEARS: Continuous Optimization for Structure Learning , 2018, NeurIPS.

[48]  Tristan Deleu,et al.  Gradient-Based Neural DAG Learning , 2019, ICLR.

[49]  Chaoxian Qi,et al.  A State-Space Approach for Detecting Stress from Electrodermal Activity , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[50]  José Manuel Gutiérrez,et al.  Who learns better Bayesian network structures: Accuracy and speed of structure learning algorithms , 2018, Int. J. Approx. Reason..

[51]  Hao Huang,et al.  Scalable Causal Graph Learning through a Deep Neural Network , 2019, CIKM.

[52]  David Maxwell Chickering,et al.  Large-Sample Learning of Bayesian Networks is NP-Hard , 2002, J. Mach. Learn. Res..

[53]  K. Hoover,et al.  Searching for the Causal Structure of a Vector Autoregression , 2003 .

[54]  Alexander J. Hartemink,et al.  Learning Non-Stationary Dynamic Bayesian Networks , 2010, J. Mach. Learn. Res..

[55]  Tom Burr,et al.  Causation, Prediction, and Search , 2003, Technometrics.