Mean Field Variational Approximation for Continuous-Time Bayesian Networks

Continuous-time Bayesian networks is a natural structured representation language for multi-component stochastic processes that evolve continuously over time. Despite the compact representation provided by this language, inference in such models is intractable even in relatively simple structured networks. We introduce a mean field variational approximation in which we use a product of inhomogeneous Markov processes to approximate a joint distribution over trajectories. This variational approach leads to a globally consistent distribution, which can be efficiently queried. Additionally, it provides a lower bound on the probability of observations, thus making it attractive for learning tasks. Here we describe the theoretical foundations for the approximation, an efficient implementation that exploits the wide range of highly optimized ordinary differential equations (ODE) solvers, experimentally explore characterizations of processes for which this approximation is suitable, and show applications to a large-scale real-world inference problem.

• Publication year

  2009

• Venue

  Journal of machine learning research

• Publication date

  2009-06-18

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.5555/1756006.1953022arXiv 1205.2655
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Continuous Time Bayesian Networks

  2012cited by this paper
• Approximate parameter inference in a stochastic reaction-diffusion model

  2010cited by this paper
• Continuous-Time Belief Propagation

  2010cited by this paper
• Parameter estimation and inference for stochastic reaction-diffusion systems: application to morphogenesis in D. melanogaster

  2010cited by this paper
• Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning

  2009cited by this paper
• Learning Continuous-Time Social Network Dynamics

  2009cited by this paper
• Gibbs Sampling in Factorized Continuous-Time Markov Processes

  2008cited by this paper
• CT-NOR: Representing and Reasoning About Events in Continuous Time

  2008cited by this paper
• Graphical Models, Exponential Families, and Variational Inference

  2008cited by this paper
• Sampling for Approximate Inference in Continuous Time Bayesian Networks

  2008cited by this paper
• Reasoning at the Right Time Granularity

  2007cited by this paper
• Variational inference for Markov jump processes

  2007influential reference
• Numerical Recipes 3rd Edition: The Art of Scientific Computing

  2007cited by this paper
• Variational Inference for Diffusion Processes

  2007cited by this paper
• Dependence among sites in RNA evolution.

  2006cited by this paper
• Residual Belief Propagation: Informed Scheduling for Asynchronous Message Passing

  2006cited by this paper
• Continuous Time Markov Networks

  2006cited by this paper
• Poisson-Networks: A Model for Structured Poisson Processes.

  2005cited by this paper
• Poisson-Networks: A Model for Structured Poisson Processes

  2005cited by this paper
• Modeling of negative autoregulated genetic networks in single cells.

  2005cited by this paper
• Expectation Maximization and Complex Duration Distributions for Continuous Time Bayesian Networks

  2005cited by this paper
• Continuous Time Particle Filtering

  2005cited by this paper
• Expectation Propagation for Continuous Time Bayesian Networks

  2005cited by this paper
• Extending Continuous Time Bayesian Networks

  2005cited by this paper
• Dynamic bayesian networks: representation, inference and learning

  2002cited by this paper
• Learning Continuous Time Bayesian Networks

  2002cited by this paper
• Tractable Inference for Complex Stochastic Processes

  1998cited by this paper
• A model for reasoning about persistence and causation

  1989cited by this paper
• Handbook of Stochastic Methods

  1983cited by this paper
• Markov Chains with Stationary Transition Probabilities

  1961cited by this paper
• Calculus of Variations

  1954cited by this paper
• On Information and Sufficiency

  1951cited by this paper
• BIOINFORMATICS ORIGINAL PAPER

  year unknowncited by this paper
• Bioinformatics Advance Access published March 11, 2009 Switching Regulatory Models of Cellular Stress Response

  year unknowncited by this paper

• Scalable Mean-Field Variational Inference via Preconditioned Primal-Dual Optimization

  2026cites this paper
• Efficient Inference for Coupled Hidden Markov Models in Continuous Time and Discrete Space

  2025cites this paper
• Structure learning for continuous time Bayesian networks via penalized likelihood

  2024cites this paper
• A Sparse Bayesian Learning for Diagnosis of Nonstationary and Spatially Correlated Faults With Application to Multistation Assembly Systems

  2023cites this paper
• Neural Markov Jump Processes

  2023cites this paper
• Entropic Matching for Expectation Propagation of Markov Jump Processes

  2023cites this paper
• Prior Density Learning in Variational Bayesian Phylogenetic Parameters Inference

  2023cites this paper
• A novel sparse Bayesian learning and its application to fault diagnosis for multistation assembly systems

  2022cites this paper
• Variational inference for Markovian queueing networks

  2021influential citation
• Markov-Modulated Hawkes Processes for Sporadic and Bursty Social Interactions

  2021influential citation
• Active learning of continuous-time Bayesian networks through interventions

  2021cites this paper
• A knowledge based system for the management of a time stamped uncertain observation set with application on preserving mobility

  2021cites this paper
• EvoLSTM: context-dependent models of sequence evolution using a sequence-to-sequence LSTM

  2020cites this paper
• Optimal Control in Partially Observable Complex Social Systems

  2020cites this paper
• Inferring a network from dynamical signals at its nodes

  2020cites this paper
• Reliability Analysis of Large-Scale Adaptive Weighted Networks

  2020cites this paper
• Propagation for Dynamic Continuous Time Chain Event Graphs

  2020cites this paper
• A Functional Model for Structure Learning and Parameter Estimation in Continuous Time Bayesian Network: An Application in Identifying Patterns of Multiple Chronic Conditions

  2020cites this paper
• Structure learning for CTBN's via penalized maximum likelihood methods

  2020cites this paper
• Moment-Based Variational Inference for Markov Jump Processes

  2019cites this paper
• Probabilistic and Bayesian Networks

  2019cites this paper
• Markov-Modulated Hawkes Processes for Sporadic and Bursty Event Occurrences

  2019influential citation
• Markov-modulated Hawkes processes for modeling sporadic and bursty event occurrences in social interactions

  2019cites this paper
• A Variational Perturbative Approach to Planning in Graph-based Markov Decision Processes

  2019cites this paper
• Point Process Models for Heterogeneous Event Time Data

  2019influential citation
• Scalable Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data

  2019influential citation
• 19 dubious ways to compute the marginal likelihood of a phylogenetic tree topology.

  2018cites this paper
• Approximate Bayesian inference with queueing networks and coupled jump processes

  2018influential citation
• Variational inequalities and mean-field approximations for partially observed systems of queueing networks

  2018influential citation
• Cluster Variational Approximations for Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data

  2018influential citation
• On the usability of continuous time bayesian networks: improving scalability and expressiveness

  2017cites this paper
• Event Detection in Continuous Video: An Inference in Point Process Approach

  2017influential citation
• Factored performance functions and decision making in continuous time Bayesian networks

  2017cites this paper
• Uncertain and negative evidence in continuous time Bayesian networks

  2016cites this paper
• Statistical physics approaches to subnetwork dynamics in biochemical systems

  2016cites this paper
• Using Social Dynamics to Make Individual Predictions: Variational Inference with a Stochastic Kinetic Model

  2016cites this paper
• Valuation and optimization for performance based logistics using continuous time Bayesian networks

  2016cites this paper
• Multivariate Continuous-Time Models: Approximate Inference Algorithms and Medical Informatics Applications

  2016cites this paper
• Path integral approaches to subnetwork description and inference

  2016cites this paper
• Sensitivity Analysis of Continuous Time Bayesian Network Reliability Models

  2015cites this paper
• Particle Gibbs algorithms for Markov jump processes

  2015cites this paper
• Unbiased Bayesian inference for population Markov jump processes via random truncations

  2015cites this paper
• Auxiliary Gibbs Sampling for Inference in Piecewise-Constant Conditional Intensity Models

  2015cites this paper
• Classification and clustering with continuous time Bayesian network models

  2015cites this paper
• Modeling Social and Temporal Context for Video Analysis

  2015influential citation
• Extensions to modeling and inference in continuous time Bayesian networks

  2014cites this paper
• Learning continuous time Bayesian network classifiers

  2014cites this paper
• Using continuous-time Bayesian networks for standards-based diagnostics and prognostics

  2014cites this paper
• Tutorial on Structured Continuous-Time Markov Processes

  2014influential citation
• Deterministic Anytime Inference for Stochastic Continuous-Time Markov Processes

  2014influential citation
• Inference Complexity in Continuous Time Bayesian Networks

  2014cites this paper
• LMMSE Estimation and Interpolation of Continuous-Time Signals from Discrete-Time Samples Using Factor Graphs

  2013cites this paper
• Formal reductions of stochastic rule-based models of biochemical systems

  2013cites this paper
• EXTENDING INFERENCE IN CONTINUOUS TIME BAYESIAN NETWORKS

  2013cites this paper
• Optimal Kullback–Leibler Aggregation via Information Bottleneck

  2013cites this paper
• Approximate model reductions for combinatorial reaction systems

  2013cites this paper
• EXTENDING INFERENCE IN CONTINUOUS TIME BAYESIAN NETWORKS by Liessman

  2013cites this paper
• Markov chain Monte Carlo for continuous-time discrete-state systems

  2012cites this paper
• Inferring divergence of context-dependent substitution rates in Drosophila genomes with applications to comparative genomics.

  2012cites this paper
• Multiplicative Forests for Continuous-Time Processes

  2012cites this paper
• Fast MCMC sampling for Markov jump processes and extensions

  2012cites this paper
• Inference in continuous-time change-point models

  2011cites this paper
• Novel models and algorithms for systems reliability modeling and optimization

  2011cites this paper
• Factored Filtering of Continuous-Time Systems

  2011influential citation
• Bayesian Time Series Models: Approximate inference for continuous-time Markov processes

  2011cites this paper
• Fast MCMC sampling for Markov jump processes and continuous time Bayesian networks

  2011cites this paper
• Reasoning about structured stochastic systems in continuous-time (עם תקציר בעברית ושער נוסף: הסקה הסתברותית במערכות דינמיות מורכבות בזמן-רציף.; שער נוסף בעברית: הסקה הסתברותית במערכות דינמיות מורכבות בזמן-רציף.)

  2011cites this paper
• Primate CpG islands are maintained by heterogeneous evolutionary regimes involving minimal selection.

  2011cites this paper
• Intrusion Detection using Continuous Time Bayesian Networks

  2010cites this paper
• A Continuous Time Bayesian Network Approach for Intrusion Detection

  2010influential citation
• Approximate inference in continuous time Gaussian-Jump processes

  2010cites this paper
• Continuous-Time Belief Propagation

  2010cites this paper
• Statistical Timeline Analysis for Electronic Health Records Table of Contents

  year unknowncites this paper