A differential approach to inference in Bayesian networks
From MaRDI portal
Publication:3452493
Abstract: We present a new approach for inference in Bayesian networks, which is mainly based on partial differentiation. According to this approach, one compiles a Bayesian network into a multivariate polynomial and then computes the partial derivatives of this polynomial with respect to each variable. We show that once such derivatives are made available, one can compute in constant-time answers to a large class of probabilistic queries, which are central to classical inference, parameter estimation, model validation and sensitivity analysis. We present a number of complexity results relating to the compilation of such polynomials and to the computation of their partial derivatives. We argue that the combined simplicity, comprehensiveness and computational complexity of the presented framework is unique among existing frameworks for inference in Bayesian networks.
Recommendations
Cited in
(59)- Portfolios in stochastic local search: efficiently computing most probable explanations in Bayesian networks
- Compiling relational Bayesian networks for exact inference
- Learning tractable Bayesian networks in the space of elimination orders
- Resolving inconsistencies of scope interpretations in sum-product networks
- Robustifying sum-product networks
- A geometric characterization of sensitivity analysis in monomial models
- Sensitivity analysis beyond linearity
- Equivalence classes of staged trees
- scientific article; zbMATH DE number 5269047 (Why is no real title available?)
- Exact stochastic constraint optimisation with applications in network analysis
- Probabilistic decision graphs for optimization under uncertainty
- Bayesian networks for discrete multivariate data: An algebraic approach to inference
- Towards an effective practice of learning from data and knowledge
- Formulating asymmetric decision problems as decision circuits
- Direct causal structure extraction from pairwise interaction patterns in NAT modeling Bayesian networks
- Quantifying the uncertainty of a belief net response: Bayesian error-bars for belief net inference
- Probabilistic decision graphs for optimization under uncertainty
- Probabilistic inference with noisy-threshold models based on a CP tensor decomposition
- Understanding the scalability of Bayesian network inference using clique tree growth curves
- scientific article; zbMATH DE number 2096817 (Why is no real title available?)
- Decision-network polynomials and the sensitivity of decision-support models
- On-line alert systems for production plants: A conflict based approach
- Fast structured prediction using large margin sigmoid belief networks
- Neural networks using Bayesian training.
- Importance sampling-based estimation over AND/OR search spaces for graphical models
- A simple graphical approach for understanding probabilistic inference in Bayesian networks
- A differential semantics for jointree algorithms
- Learning to assign degrees of belief in relational domains
- Logical Compilation of Bayesian Networks with Discrete Variables
- Efficient algorithms for robustness analysis of maximum a posteriori inference in selective sum-product networks
- On converting sum-product networks into Bayesian networks
- Case-factor diagrams for structured probabilistic modeling
- A Differential Approach for Staged Trees
- A simplified matrix formulation for sensitivity analysis of hidden Markov models
- An extension of the differential approach for Bayesian network inference to dynamic Bayesian networks
- Discovery of statistical equivalence classes using computer algebra
- Knowledge compilation. Preface.
- Sensitivity analysis in multilinear probabilistic models
- Connecting knowledge compilation classes and width parameters
- On probabilistic inference by weighted model counting
- Efficient sensitivity analysis in hidden Markov models
- SampleSearch: importance sampling in presence of determinism
- \textsc{Strudel}: A fast and accurate learner of structured-decomposable probabilistic circuits
- Conditional sum-product networks: modular probabilistic circuits via gate functions
- A symbolic algebra for the computation of expected utilities in multiplicative influence diagrams
- On the relative expressiveness of Bayesian and neural networks
- Learning multi-linear representations of distributions for efficient inference
- A hierarchy of sum-product networks using robustness
- Sum-product graphical models
- Structural extension to logistic regression: Discriminative parameter learning of belief net classifiers
- Coherent combination of probabilistic outputs for group decision making: an algebraic approach
- AND/OR search spaces for graphical models
- scientific article; zbMATH DE number 1149441 (Why is no real title available?)
- Learning directed acyclic graph SPNs in sub-quadratic time
- The YODO algorithm: an efficient computational framework for sensitivity analysis in Bayesian networks
- Visualizing and understanding sum-product networks
- The complexity of Bayesian networks specified by propositional and relational languages
- Connecting width and structure in knowledge compilation
- Discriminative training of feed-forward and recurrent sum-product networks by extended Baum-Welch
This page was built for publication: A differential approach to inference in Bayesian networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3452493)
