Temporal Predictions with Bayesian Compositional Hierarchies
نویسنده
چکیده
In this note I describe a novel approach to modelling and exploiting probabilistic dependencies in compositional hierarchies for model-based scene interpretation. I present Bayesian Compositional Hierarchies (BCHs) which capture all probabilistic information about the objects of a compositional hierarchy in object-centered aggregate representations. BCHs extend typical Bayesian Network models by allowing arbitrary probabilistic dependencies within aggregates, yet providing efficient inference procedures. New closed-form solutions are presented for inferences in a multivariate Gaussian BCH. Results are presented comparing a BCH with existing methods (pure Bayesian Networks, unrestricted Joint Probability Distributions). Monitoring aircraft service operations is presented as a practical application. It is shown that predictions about the expected temporal development of service operations can be generated dynamically from available temporal data. Temporal Predictions with Bayesian Compositional Hierarchies
منابع مشابه
Bayesian Compositional Hierarchies - A Probabilistic Structure for Scene Interpretation
In high-level vision, it is often useful to organize conceptual models in compositional hierarchies. For example, models of building facades (which are used here as examples) can be described in terms of constituent parts such as balconies or window arrays which in turn may be further decomposed. While compositional hierarchies are widely used in scene interpretation, it is not clear how to mod...
متن کاملContext-Based Probabilistic Scene Interpretation
In high-level scene interpretation, it is useful to exploit the evolving probabilistic context for stepwise interpretation decisions. We present a new approach based on a general probabilistic framework and beam search for exploring alternative interpretations. As probabilistic scene models, we propose Bayesian Compositional Hierarchies (BCHs) which provide object-centered representations of co...
متن کاملAnalysis of Hierarchical Bayesian Models for Large Space Time Data of the Housing Prices in Tehran
Housing price data is correlated to their location in different neighborhoods and their correlation is type of spatial (location). The price of housing is varius in different months, so they also have a time correlation. Spatio-temporal models are used to analyze this type of the data. An important purpose of reviewing this type of the data is to fit a suitable model for the spatial-temporal an...
متن کاملLearning of Compositional Hierarchies By Data-Driven Chunking
Compositional hierarchies (CHs), layered structures of part-of relationships, underlie many forms of data, and representations involving these structures lie at the heart of much of AI. Despite this importance, methods for learning CHs from data are scarce. We present an unsupervised technique for learning CHs by an on-line, bottom-up chunking process. At any point, the induced structure can ma...
متن کاملDeveloping a Compositional Reservoir Model for Investigating the Effect of Interfacial Tension on Oil Recovery
In this paper, a simplified formulation for compositional reservoir simulator is presented. These types of simulators are used when inter-phase mass transfer depends on phase composition as well as pressure. The procedure for solving compositional model equations is completely described. For equilibrium calculation, property estimation Peng Robinson equation of state is used. This equation ...
متن کامل