The Experts below are selected from a list of 19431 Experts worldwide ranked by ideXlab platform
Karl Henrik Johansson - One of the best experts on this subject based on the ideXlab platform.
-
The Role of Persistent Graphs in the Agreement Seeking of Social Networks
IEEE Journal on Selected Areas in Communications, 2013Co-Authors: Guodong Shi, Karl Henrik JohanssonAbstract:This paper investigates the role persistent relations play for a social network to reach a Global Belief agreement under discrete-time or continuous-time evolution. Each directed arc in the underlying communication graph is assumed to be associated with a time-dependent weight function, which describes the strength of the information flow from one node to another. An arc is said to be persistent if its weight function has infinite L1 or l1 norm for continuous or discrete Belief evolutions, respectively. The graph that consists of all persistent arcs is called the persistent graph of the underlying network. Three necessary and sufficient conditions on agreement or e-agreement are established. We prove that the persistent graph fully determines the convergence to a common opinion in a social network. It is shown how the convergence rate explicitly depends on the diameter of the persistent graph. For a social networking service like Facebook, our results indicate how permanent friendships need to be and what network topology they should form for the network to be an efficient platform for opinion diffusion.
-
The Role of Persistent Graphs in the Agreement Seeking of Social Networks
arXiv: Multiagent Systems, 2011Co-Authors: Guodong Shi, Karl Henrik JohanssonAbstract:This paper investigates the role persistent arcs play for a social network to reach a Global Belief agreement under discrete-time or continuous-time evolution. Each (directed) arc in the underlying communication graph is assumed to be associated with a time-dependent weight function which describes the strength of the information flow from one node to another. An arc is said to be persistent if its weight function has infinite $\mathscr{L}_1$ or $\ell_1$ norm for continuous-time or discrete-time Belief evolutions, respectively. The graph that consists of all persistent arcs is called the persistent graph of the underlying network. Three necessary and sufficient conditions on agreement or $\epsilon$-agreement are established, by which we prove that the persistent graph fully determines the convergence to a common opinion in social networks. It is shown how the convergence rates explicitly depend on the diameter of the persistent graph. The results adds to the understanding of the fundamentals behind Global agreements, as it is only persistent arcs that contribute to the convergence.
Adnan Darwiche - One of the best experts on this subject based on the ideXlab platform.
-
Sensitivity analysis of probabilistic graphical models
2005Co-Authors: Hei Chan, Adnan DarwicheAbstract:Probabilistic Belief systems are used in artificial intelligence to model uncertainty. A popular framework for realizing probabilistic Belief systems is to use graphical models, such as Bayesian networks and Markov networks. The topic of sensitivity analysis is concerned broadly with the relationships between local Beliefs, such as network parameters, and Global Beliefs, such as values of probabilistic queries. Sensitivity analysis is crucial to probabilistic Belief systems because we often need to revise our state of Belief to incorporate new probabilistic information in the form of local Belief changes. This work focuses on sensitivity analysis of probabilistic graphical models, by addressing central research problems such as the assessment of Global Belief changes due to local Belief changes, the identification of local Belief changes that induce certain Global Belief changes, and the quantifying of Belief changes in general. Our results can be divided into the following parts. First, we develop procedures and complexity results for tuning Bayesian or Markov network parameters (single or multiple) to ensure certain query constraints. Second, we provide network-independent bounds on changes in query values due to arbitrary changes in Bayesian or Markov network parameters. Third, we propose a new distance measure for quantifying probabilistic Belief changes, and use it to provide guarantees on Global Belief changes in Bayesian or Markov networks. Fourth, we provide algorithms and complexity results on the sensitivity of decisions induced by Bayesian networks. Finally, we discuss the philosophical topic of Belief revision. Many of our results have been implemented in a program called SamIam (Sensitivity Analysis, Modeling, Inference and More), a graphical Bayesian network tool developed by the UCLA Automated Reasoning Group.
-
A distance measure for bounding probabilistic Belief change
International Journal of Approximate Reasoning, 2005Co-Authors: Hei Chan, Adnan DarwicheAbstract:We propose a distance measure between two probability distributions, which allows one to bound the amount of Belief change that occurs when moving from one distribution to another. We contrast the proposed measure with some well known measures, including KL-divergence, showing some theoretical properties on its ability to bound Belief changes. We then present two practical applications of the proposed distance measure: sensitivity analysis in Belief networks and probabilistic Belief revision. We show how the distance measure can be easily computed in these applications, and then use it to bound Global Belief changes that result from either the perturbation of local conditional Beliefs or the accommodation of soft evidence. Finally, we show that two well known techniques in sensitivity analysis and Belief revision correspond to the minimization of our proposed distance measure and, hence, can be shown to be optimal from that viewpoint.
-
AAAI/IAAI - A distance measure for bounding probabilistic Belief change
2002Co-Authors: Hei Chan, Adnan DarwicheAbstract:We propose a distance measure between two probability distributions, which allows one to bound the amount of Belief change that occurs when moving from one distribution to another. We contrast the proposed measure with some well known measures, including KL-divergence, showing how they fail to be the basis for bounding Belief change as is done using the proposed measure. We then present two practical applications of the proposed distance measure: sensitivity analysis in Belief networks and probabilistic Belief revision. We show how the distance measure can be easily computed in these applications, and then use it to bound Global Belief changes that result from either the perturbation of local conditional Beliefs or the accommodation of soft evidence. Finally, we show that two well known techniques in sensitivity analysis and Belief revision correspond to the minimization of our proposed distance measure and, hence, can be shown to be optimal from that viewpoint.
Hei Chan - One of the best experts on this subject based on the ideXlab platform.
-
Sensitivity Analysis of Probabilistic Graphical Models: Theoretical Results and Their Applications on Bayesian Network Modeling and Inference
2009Co-Authors: Hei ChanAbstract:Probabilistic graphical models such as Bayesian networks are widely used for large-scale data analysis in various fields such as customer data analysis and medical diagnosis, as they model probabilistic knowledge naturally and allow the use of efficient inference algorithms to draw conclusions from the model. Sensitivity analysis of probabilistic graphical models is the analysis of the relationships between the inputs (local Beliefs), such as network parameters, and the outputs (Global Beliefs), such as values of probabilistic queries, and addresses the central research problem of how Beliefs will be changed when we incorporate new information to the current model. This book provides many theoretical results, such as the assessment of Global Belief changes due to local Belief changes, the identification of local Belief changes that induce certain Global Belief changes, and the quantifying of Belief changes in general. These results can be applied on the modeling and inference of Bayesian networks, and provide a critical tool for the researchers, developers, and users of Bayesian networks during the process of probabilistic data modeling and reasoning.
-
Sensitivity analysis of probabilistic graphical models
2005Co-Authors: Hei Chan, Adnan DarwicheAbstract:Probabilistic Belief systems are used in artificial intelligence to model uncertainty. A popular framework for realizing probabilistic Belief systems is to use graphical models, such as Bayesian networks and Markov networks. The topic of sensitivity analysis is concerned broadly with the relationships between local Beliefs, such as network parameters, and Global Beliefs, such as values of probabilistic queries. Sensitivity analysis is crucial to probabilistic Belief systems because we often need to revise our state of Belief to incorporate new probabilistic information in the form of local Belief changes. This work focuses on sensitivity analysis of probabilistic graphical models, by addressing central research problems such as the assessment of Global Belief changes due to local Belief changes, the identification of local Belief changes that induce certain Global Belief changes, and the quantifying of Belief changes in general. Our results can be divided into the following parts. First, we develop procedures and complexity results for tuning Bayesian or Markov network parameters (single or multiple) to ensure certain query constraints. Second, we provide network-independent bounds on changes in query values due to arbitrary changes in Bayesian or Markov network parameters. Third, we propose a new distance measure for quantifying probabilistic Belief changes, and use it to provide guarantees on Global Belief changes in Bayesian or Markov networks. Fourth, we provide algorithms and complexity results on the sensitivity of decisions induced by Bayesian networks. Finally, we discuss the philosophical topic of Belief revision. Many of our results have been implemented in a program called SamIam (Sensitivity Analysis, Modeling, Inference and More), a graphical Bayesian network tool developed by the UCLA Automated Reasoning Group.
-
A distance measure for bounding probabilistic Belief change
International Journal of Approximate Reasoning, 2005Co-Authors: Hei Chan, Adnan DarwicheAbstract:We propose a distance measure between two probability distributions, which allows one to bound the amount of Belief change that occurs when moving from one distribution to another. We contrast the proposed measure with some well known measures, including KL-divergence, showing some theoretical properties on its ability to bound Belief changes. We then present two practical applications of the proposed distance measure: sensitivity analysis in Belief networks and probabilistic Belief revision. We show how the distance measure can be easily computed in these applications, and then use it to bound Global Belief changes that result from either the perturbation of local conditional Beliefs or the accommodation of soft evidence. Finally, we show that two well known techniques in sensitivity analysis and Belief revision correspond to the minimization of our proposed distance measure and, hence, can be shown to be optimal from that viewpoint.
-
AAAI/IAAI - A distance measure for bounding probabilistic Belief change
2002Co-Authors: Hei Chan, Adnan DarwicheAbstract:We propose a distance measure between two probability distributions, which allows one to bound the amount of Belief change that occurs when moving from one distribution to another. We contrast the proposed measure with some well known measures, including KL-divergence, showing how they fail to be the basis for bounding Belief change as is done using the proposed measure. We then present two practical applications of the proposed distance measure: sensitivity analysis in Belief networks and probabilistic Belief revision. We show how the distance measure can be easily computed in these applications, and then use it to bound Global Belief changes that result from either the perturbation of local conditional Beliefs or the accommodation of soft evidence. Finally, we show that two well known techniques in sensitivity analysis and Belief revision correspond to the minimization of our proposed distance measure and, hence, can be shown to be optimal from that viewpoint.
Christopher D Manning - One of the best experts on this subject based on the ideXlab platform.
-
Global Belief recursive neural networks
Neural Information Processing Systems, 2014Co-Authors: Romain Paulus, Richard Socher, Christopher D ManningAbstract:Recursive Neural Networks have recently obtained state of the art performance on several natural language processing tasks. However, because of their feedforward architecture they cannot correctly predict phrase or word labels that are determined by context. This is a problem in tasks such as aspect-specific sentiment classification which tries to, for instance, predict that the word Android is positive in the sentence Android beats iOS. We introduce Global Belief recursive neural networks (GB-RNNs) which are based on the idea of extending purely feedforward neural networks to include one feedbackward step during inference. This allows phrase level predictions and representations to give feedback to words. We show the effectiveness of this model on the task of contextual sentiment analysis. We also show that dropout can improve RNN training and that a combination of unsupervised and supervised word vector representations performs better than either alone. The feedbackward step improves F1 performance by 3% over the standard RNN on this task, obtains state-of-the-art performance on the SemEval 2013 challenge and can accurately predict the sentiment of specific entities.
-
NIPS - Global Belief Recursive Neural Networks
2014Co-Authors: Romain Paulus, Richard Socher, Christopher D ManningAbstract:Recursive Neural Networks have recently obtained state of the art performance on several natural language processing tasks. However, because of their feedforward architecture they cannot correctly predict phrase or word labels that are determined by context. This is a problem in tasks such as aspect-specific sentiment classification which tries to, for instance, predict that the word Android is positive in the sentence Android beats iOS. We introduce Global Belief recursive neural networks (GB-RNNs) which are based on the idea of extending purely feedforward neural networks to include one feedbackward step during inference. This allows phrase level predictions and representations to give feedback to words. We show the effectiveness of this model on the task of contextual sentiment analysis. We also show that dropout can improve RNN training and that a combination of unsupervised and supervised word vector representations performs better than either alone. The feedbackward step improves F1 performance by 3% over the standard RNN on this task, obtains state-of-the-art performance on the SemEval 2013 challenge and can accurately predict the sentiment of specific entities.
Guodong Shi - One of the best experts on this subject based on the ideXlab platform.
-
The Role of Persistent Graphs in the Agreement Seeking of Social Networks
IEEE Journal on Selected Areas in Communications, 2013Co-Authors: Guodong Shi, Karl Henrik JohanssonAbstract:This paper investigates the role persistent relations play for a social network to reach a Global Belief agreement under discrete-time or continuous-time evolution. Each directed arc in the underlying communication graph is assumed to be associated with a time-dependent weight function, which describes the strength of the information flow from one node to another. An arc is said to be persistent if its weight function has infinite L1 or l1 norm for continuous or discrete Belief evolutions, respectively. The graph that consists of all persistent arcs is called the persistent graph of the underlying network. Three necessary and sufficient conditions on agreement or e-agreement are established. We prove that the persistent graph fully determines the convergence to a common opinion in a social network. It is shown how the convergence rate explicitly depends on the diameter of the persistent graph. For a social networking service like Facebook, our results indicate how permanent friendships need to be and what network topology they should form for the network to be an efficient platform for opinion diffusion.
-
The Role of Persistent Graphs in the Agreement Seeking of Social Networks
arXiv: Multiagent Systems, 2011Co-Authors: Guodong Shi, Karl Henrik JohanssonAbstract:This paper investigates the role persistent arcs play for a social network to reach a Global Belief agreement under discrete-time or continuous-time evolution. Each (directed) arc in the underlying communication graph is assumed to be associated with a time-dependent weight function which describes the strength of the information flow from one node to another. An arc is said to be persistent if its weight function has infinite $\mathscr{L}_1$ or $\ell_1$ norm for continuous-time or discrete-time Belief evolutions, respectively. The graph that consists of all persistent arcs is called the persistent graph of the underlying network. Three necessary and sufficient conditions on agreement or $\epsilon$-agreement are established, by which we prove that the persistent graph fully determines the convergence to a common opinion in social networks. It is shown how the convergence rates explicitly depend on the diameter of the persistent graph. The results adds to the understanding of the fundamentals behind Global agreements, as it is only persistent arcs that contribute to the convergence.