As computer architectures move towards multicore we must build a theoretical understanding of parallelism in machine learning. In this paper we focus on parallel inference in graphical models. We demonstrate that the natural, fully synchronous parallelization of belief propagation is highly inefficient. By bounding the achievable parallel performance in chain graphical models we develop a theor...

Perhaps the most popular approach to animating algorithms consists of identifying interesting events in the implementation code, corresponding to relevant actions in the underlying algorithm, and turning them into graphical events by inserting calls to suitable visualization routines. Another natural approach conceives algorithm animation as a graphical interpretation of the state of the comput...

Recently, a class of multiscale tree-structured models was introduced in terms of scale-recursive dynamics deened on trees. The main advantage of these models is their association with a fast, recursive, Kalman-lter prediction algorithm. In this article, we propose a more general class of multiscale graphical models over acyclic directed graphs, for use in command and control problems. Moreover...

We address the inference of discrete-state models for tree-structured data. Our aim is to introduce parametric multitype branching processes that can be efficiently estimated on the basis of data of limited size. Each generation distribution within this macroscopic model is modeled by a partially directed acyclic graphical model. The estimation of each graphical model relies on a greedy algorit...

Collective graphical models (CGMs) are a formalism for inference and learning with aggregate data that are motivated by a model for bird migration. We highlight a close connection between approximate MAP inference in CGMs and marginal inference in standard graphical models. The connection leads us to derive a novel Belief Propagation (BP)-style algorithm for collective graphical models. The alg...

Tracking by sequential Bayesian filtering relies on a graphical model with temporally ordered linear structure based on temporal smoothness assumption. This framework is convenient to propagate the posterior through the first-order Markov chain. However, density propagation from a single immediately preceding frame may be unreliable especially in challenging situations such as abrupt appearance...

The paper introduces a generalization for known probabilistic models such as log-linear and graphical models, called here multiplicative models. These models, that express probabilities via product of parameters are shown to capture multiple forms of contextual independence between variables, including decision graphs and noisy-OR functions. An inference algorithm for multiplicative models is p...

In reality there are many relational datasets in which both features of instances and the relationships among the instances are recorded, such as hyperlinked web pages, scientific literature with citations, and social networks. Collective classification has been widely used to classify a group of related instances simultaneously. Recently there have been several studies on statistical relationa...