Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning

By Daphne Koller, Nir Friedman. 2009

8,114 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

Citations in the corpus

Diffusion-Convolutional Neural Networks (2015, background): “Probabilistic Relational Models DCNNs also share strong ties to probabilistic relational models (PRMs), a family of graphical models that are capable of representing distributions over relational data [16].”
Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Review (2018, background): “Probabilistic graphical models (PGMs) offer a broadly applicable and useful toolbox for the machine learning researcher (Koller and Friedman, 2009): by couching the entirety of the learning problem in the parlance of probability theory, they provide a consistent and flexible framework to devise…”
Bayesian Structure Learning with Generative Flow Networks (2022, methodology): “This can be explained by the fact that with a larger amount of data, the posterior distribution becomes very peaky (Koller and Friedman, 2009).”
Gradient-Based Neural DAG Learning (2019, methodology): “In this work, we propose a novel score-based method [5, 12] for structure learning named GraN-DAG which makes use of a recent reformulation of the original combinatorial problem of finding an optimal DAG into a continuous constrained optimization problem.”
RNADE: The real-valued neural autoregressive density-estimator (2013, methodology): “In previous experiments, Zoran and Weiss [3] subtracted the mean pixel value from each patch, reducing the dimensionality of the data by one: the value of any pixel could be perfectly predicted as minus the sum of all other pixel values.”
Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation (2015, methodology): “ucture, hence we need to apply approximate inference algorithm to solve the above MAP problem. Here we have applied an efﬁcient message passing algorithm which is based on the mean ﬁeld approximation [24], [14]. The mean ﬁeld approximation constructs a simpler distribution QQ(y), e.g., a product of ”
Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials (2011, background): “Instead of computing the exact distribution P (X), the mean field approximation computes a distribution Q(X) that minimizes the KL-divergence D(Q‖P ) among all distributions Q that can be expressed as a product of independent marginals, Q(X) = ∏ iQi(Xi) [10].”
A Survey of Available Corpora For Building Data-Driven Dialogue Systems: The Journal Version (2015, background): “There exist certain functions fi, for which exact inference is tractable and learning the parameters θ is efficient (Koller and Friedman, 2009; Serban, 2012).”
Fully Connected Deep Structured Networks (2015, background): “Due to non-convexity, only convergence to a stationary point of the KL divergence cost function is guaranteed for sequential block-coordinate updates [38, 16].”
Stochastic variational inference (2012, methodology): “In particular, probabilistic graphical models give us a visual language for expressing assumptions about data and its hidden structure.”
Learning Deep Structured Models (2014, methodology): “In the future we plan to learn deeper models for object recognition in applications such as holistic semantic scene understanding.”
Counterfactual Data Augmentation for Mitigating Gender Stereotypes in Languages with Rich Morphology (2019, methodology): “In this section, we present a Markov random field (MRF; Koller and Friedman, 2009) for morpho-”
Composing graphical models with neural networks for structured representations and fast inference (2016, methodology): “Probabilistic graphical models [1, 2] provide many tools to build structured representations, but often make rigid assumptions and may require significant feature engineering.”
Causal Modeling-Based Discrimination Discovery and Removal: Criteria, Bounds, and Algorithms (2019, methodology): “Under this assumption, the joint distribution over all attributes P ðvÞ can be computed using the factorization formula [26]”
A Causal Framework for Discovering and Removing Direct and Indirect Discrimination (2016, methodology): “The joint distribution over all attributes P ( v ) can be computed using the factorization formula (Koller and Friedman 2009) where P ( v | Pa ( V )) is the observational distribution.”
Achieving Non-Discrimination in Data Release (2016, methodology): “The joint probability distribution can be computed using the factorization formula [19]:”
A Comprehensive Survey of Neural Architecture Search (2020, background): “These local decisions caused by the bias problem in graph-based models [80, 81] will limit DARTS to make the optimal architectural choices.”