Belief Propagation (BP) is a widely used approximation for exact probabilistic inference in graphical models, such as Markov Random Fields (MRFs). In graphs with cycles, however, no exact convergence guarantees for BP are known, in general. For the case when all edges in the MRF carry the same symmetric, doubly stochastic potential, recent works have proposed to approximate BP by linearizing the update equations around default values, which was shown to work well for the problem of node classification. The present paper generalizes all prior work and derives an approach that approximates loopy BP on any pairwise MRF with the problem of solving a linear equation system. This approach combines exact convergence guarantees and a fast matrix implementation with the ability to model heterogenous networks. Experiments on synthetic graphs with planted edge potentials show that the linearization has comparable labeling accuracy as BP for graphs with weak potentials, while speeding-up inference by orders of magnitude.

Next week we're going to discuss the very controversial awards season contender, Nate Parker's The Birth of a Nation. Are you planning to see it in theaters? Record and send us a voice memo at slaterepresent@gmail.com or leave us a message at 646-580-1748 and your thoughts might get shared on next week's episode.

Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Recent years have seen a growing interest in player modeling, which supports the creation of player-adaptive digital games. A central problem of player modeling is goal recognition, which aims to recognize players’ intentions from observable gameplay behaviors. Player goal recognition offers the promise of enabling games to dynamically adjust challenge levels, perform procedural content generation, and create believable NPC interactions. A growing body of work is investigating a wide range of machine learning-based goal recognition models. In this paper, we introduce GOALIE, a multidimensional framework for evaluating player goal recognition models. The framework integrates multiple metrics for player goal recognition models, including two novel metrics, n-early convergence rate and standardized convergence point . We demonstrate the application of the GOALIE framework with the evaluation of several player goal recognition models, including Markov logic network-based, deep feedforward neural network-based, and long short-term memory network-based goal recognizers on two different educational games. The results suggest that GOALIE effectively captures goal recognition behaviors that are key to next-generation player modeling.

In this paper, we consider a variant of belief propagation algorithm in a tree graphical model where computations are carried out in the negative log-likelihood domain. Unlike the min-product algorithm, our goal is not limited to estimating the mode of the marginal distribution. We would like to obtain the entire marginal distribution as the sum-product algorithm does. We applied the algorithm to learn effective users features for A/B testing. We discussed scalable extension to the proposed algorithm for processing large amount of data.The primary goal of a parallel program is to reduce running time comparing to the sequential program by taking full advantage of computing power of multiprocessors.

Two essential tasks in managing description logic knowledge bases are eliminating problematic axioms and incorporating newly formed ones. Such elimination and incorporation are formalised as the operations of contraction and revision in belief change. In this paper, we deal with contraction and revision for the DL-Lite family through a model-theoretic approach. Standard description logic semantics yields an infinite number of models for DL-Lite knowledge bases, thus it is difficult to develop algorithms for contraction and revision that involve DL models. The key to our approach is the introduction of an alternative semantics called type semantics which can replace the standard semantics in characterising the standard inference tasks of DL-Lite.

Two essential tasks in managing description logic knowledge bases are eliminating problematic axioms and incorporating newly formed ones. Such elimination and incorporation are formalised as the operations of contraction and revision in belief change. In this paper, we deal with contraction and revision for the DL-Lite family through a model-theoretic approach. Standard description logic semantics yields an infinite number of models for DL-Lite knowledge bases, thus it is difficult to develop algorithms for contraction and revision that involve DL models. The key to our approach is the introduction of an alternative semantics called type semantics which can replace the standard semantics in characterising the standard inference tasks of DL-Lite. Type semantics has several advantages over the standard one. It is more succinct and importantly, with a finite signature, the semantics always yields a finite number of models. We then define model-based contraction and revision functions for DL-Lite knowledge bases under type semantics and provide representation theorems for them. Finally, the finiteness and succinctness of type semantics allow us to develop tractable algorithms for instantiating the functions.

Abduction is a pattern of inference in which an agent seeks an explanation for an observation or report. Iterated abduction is a variety of abduction in which evidence is acquired and explained over time. The long-term goal is to maintain highly plausible consistent explanations for as much of the evidence as possible. Some reports, at the time of acquisition, may be inconsistent with the agent's present beliefs, so some beliefs must be contracted in order to find an explanation of the new reports. Existing work in iterated belief revision only addresses how to maintain consistent beliefs in light of inconsistent observations; whether or not existing beliefs serve as explanations is not considered. What is needed to meet the goal of iterated abduction is a means of seeking new explanations for old evidence when previously-accepted explanations are contracted. We develop a logical formalism for this process as well as a computational implementation.

Merging beliefs requires the plausibility of the sources of the information to be merged. They are typically assumed equally reliable in lack of hints indicating otherwise; yet, a recent line of research spun from the idea of deriving this information from the revision process itself. In particular, the history of previous revisions and previous merging examples provide information for performing subsequent mergings. Yet, no examples or previous revisions may be available. In spite of the apparent lack of information, something can still be inferred by a try-and-check approach: a relative reliability ordering is assumed, the merging process is performed based on it, and the result is compared with the original information. The outcome of this check may be incoherent with the initial assumption, like when a completely reliable source is rejected some of the information it provided. In such cases, the reliability ordering assumed in the first place can be excluded from consideration. The first theorem of this article proves that such a scenario is indeed possible. Other results are obtained under various definition of reliability and merging.

Dynamic Bayesian networks (DBNs) are a general model for stochastic processes with partially observed states. Belief filtering in DBNs is the task of inferring the belief state (i.e. the probability distribution over process states) based on incomplete and noisy observations. This can be a hard problem in complex processes with large state spaces. In this article, we explore the idea of accelerating the filtering task by automatically exploiting causality in the process. We consider a specific type of causal relation, called passivity, which pertains to how state variables cause changes in other variables. We present the Passivity-based Selective Belief Filtering (PSBF) method, which maintains a factored belief representation and exploits passivity to perform selective updates over the belief factors. PSBF produces exact belief states under certain assumptions and approximate belief states otherwise, where the approximation error is bounded by the degree of uncertainty in the process. We show empirically, in synthetic processes with varying sizes and degrees of passivity, that PSBF is faster than several alternative methods while achieving competitive accuracy. Furthermore, we demonstrate how passivity occurs naturally in a complex system such as a multi-robot warehouse, and how PSBF can exploit this to accelerate the filtering task.