• Corpus ID: 14995779

Probabilistic Graphical Models - Principles and Techniques

  title={Probabilistic Graphical Models - Principles and Techniques},
  author={Daphne Koller and Nir Friedman},
Most tasks require a person or an automated system to reason -- to reach conclusions based on available information. The framework of probabilistic graphical models, presented in this book, provides a general approach for this task. The approach is model-based, allowing interpretable models to be constructed and then manipulated by reasoning algorithms. These models can also be learned automatically from data, allowing the approach to be used in cases where manually constructing a model is… 
Graphical models and symmetries: loopy belief propagation approaches
This thesis deals with efficient inference exploiting symmetries in graphical models for various query types and introduces lifted loopy belief propagation (lifted LBP), the first lifted parallel inference approach for relational as well as propositional graphical models.
Introduction to Probabilistic Graphical Models
Probabilistic Graphical Models
  • S. Srihari
  • Computer Science
    Encyclopedia of Social Network Analysis and Mining
  • 2014
This course will provide a comprehensive survey of state-of-the-art methods for statistical learning and inference in graphical models, including variational methods, which adapt tools from optimization theory to develop efficient, possibly approximate, inference algorithms.
Reasoning and Decisions in Probabilistic Graphical Models - A Unified Framework
This thesis proposes a spectrum of efficient belief propagation style algorithms with "message passing" forms, which are simple, fast and amenable to parallel or distributed computation, and derives a class of efficient algorithms that combines the advantages of several existing algorithms, resulting in improved performance on traditional marginalization and optimization tasks.
A natural extension of the ordinary Markov approach is described, whereby both conditional independences and generalized constraints are used to define a nested Markov model, and most structural features of hidden variable DAGs can be recovered exactly when a single generalized independence constraint holds under the distribution of the observed variables.
Methods for Learning Directed and Undirected Graphical Models
A non-parametric method for learning undirected graphs from continuous data which combines a conditional mutual information estimator with a permutation test in order to perform conditional independence testing without assuming any specific parametric distributions for the involved random variables.
A Probabilistic Programming Approach To Probabilistic Data Analysis
This paper introduces composable generative population models (CGPMs), a computational abstraction that extends directed graphical models and can be used to describe and compose a broad class of probabilistic data analysis techniques.
Query-Specific Learning and Inference for Probabilistic Graphical Models
A polynomial time algorithm is proposed for learning the structure of tractable graphical models with quality guarantees, including PAC learnability and graceful degradation guarantees, which is the first efficient algorithm to provide this type of guarantees.
Belief Graphical Models for Uncertainty Representation and Reasoning
This chapter provides an overview of the most common belief graphical models and gives an overview on various aspects related to graphical models for uncertainty: representation, inference, learning and finally applications.


Introduction discussion of some of these alternative frameworks see Shafer and Pearl
  • Introduction discussion of some of these alternative frameworks see Shafer and Pearl
  • 1988