Probabilistic reasoning in intelligent systems - networks of plausible inference

By Judea Pearl. 1989

Subfields: Probabilistic & Bayesian Methods, Knowledge Graphs & Representation

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

In Knowledge Graphs & Representation

Citations in the corpus

Discriminative Embeddings of Latent Variable Models for Structured Data (2016, positive): “This setting is similar to Pyzer-Knapp et al. (2015), except that we use the entire 2.”
Natural Language Processing (Almost) from Scratch (2011, background): “Separately training a submodel only ar Xi v makes sense when the training data blocks these additional dependency paths (in the sense of d-separation, Pearl, 1988).”
Variational Inference: A Review for Statisticians (2016, methodology): “Variational message passing connects variational inference to the classical theories of graphical models and probabilistic inference (Pearl 1988; Lauritzen and Spiegelhalter 1988).”
Hierarchical Reinforcement Learning with the MAXQ Value Function Decomposition (1999, methodology): “Our philosophy in developing MAXQ (which we share with other reinforcement learning researchers, notably Parr and Russell) has been to draw inspiration from the development of Belief Networks (Pearl, 1988).”
Active Learning with Statistical Models (1996, background): “Of particular interest is the class of models known as \belief networks" or \Bayesian networks" (Pearl, 1988; Heckerman, Geiger, & Chickering, 1994).”
Mean Field Theory for Sigmoid Belief Networks (1996, methodology): “This e(cid:11)ective input is composed of terms from the unit's Markov blanket (Pearl, 1988), shown in Figure 3; in particular, these terms take into account the unit's internal bias, the values of its parents and children, and, through the matrix Kji, the values of its children's other parents.”
Learning Bayesian networks: The combination of knowledge and statistical data (1994, background): “A Bayesian network is an annotated directed graph that encodes probabilistic relationships among distinctions of interest in an uncertain-reasoning problem (Howard & Matheson, 1981; Pearl, 1988).”
Strong completeness and faithfulness in Bayesian networks (1995, methodology): “Pearl's claim is supported by the proof that for any di rected acyclic graph G there exists distribution for which G is a perfect map (o r, in the language of this paper, the distribution is faithful to G); see Theorem 10 of Pearl1988 and Geiger and Pearl1988b for a proof.”
Inductive biases for deep learning of higher-level cognition (2020, background): “Causal models specify how the joint distribution of interest could change under interventions (Pearl, 1988), or more generally actions (Sutton and Barto, 2018).”
Gradient-based boosting for statistical relational learning: The relational dependency network case (2011, background): “Bayesian networks (BNs) (Pearl 1988) have been widely used for learning and reasoning about probabilistic relationships.”
Effective Bayesian Inference for Stochastic Programs (1997, background): “The different requirements of these tasks have resulted in the use of different stochastic modeling languages, such as Bayesian networks (Pearl 1988) and dynamic Bayesian networks (Dean and Kanazawa 1989), hidden Markov models (Rabiner and Juang 1986), and stochastic context-free grammars (SCFGs)…”
Top-down and bottom-up cues for scene text recognition (2012, methodology): “These distributions are then used to reweight the messages being passed during loopy BP [24] on each tree.”
Near optimum universal belief propagation based decoding of low-density parity check codes (2002, methodology): “However, soft decoding achieves a much better performance, and can be implemented by iterative decoding based on the belief propagation (BP) 1 algorithm [ 12 ].”
Learning Bayesian networks: approaches and issues (2011, methodology): “Examples can be found in the paper by Lauritzen and Wermuth (1989) and the books by Castillo et al. (1997a), Pearl (1988) and Whittaker (1990).”
Tree consistency and bounds on the performance of the max-product algorithm and its generalizations (2004, methodology): “The perspective that we take differs from other treatments (e.g., Pearl 1988), in that we interpret max-product as a particular method for computing the max-marginal reparameterization of Theorem 1.”