Tree-Structured Stick Breaking Processes for Hierarchical Data

Many data are naturally modeled by an unobserved hierarchical structure. In this paper we propose a flexible nonparametric prior over unknown data hierarchies. The approach uses nested stick-breaking processes to allow for trees of unbounded width and depth, where data can live at any node and are infinitely exchangeable. One can view our model as providing infinite mixtures where the components have a dependency structure corresponding to an evolutionary diffusion down a tree. By using a stick-breaking approach, we can apply Markov chain Monte Carlo methods based on slice sampling to perform Bayesian inference and simulate from the posterior distribution on trees. We apply our method to hierarchical clustering of images and topic modeling of text data.

Analysis of the piston ring/liner oil film development during warm-up for an SI-engine

A one-dimensional ring-pack lubrication model developed at MIT is applied to simulate the oil film behavior during the warm-up period of a Kohler spark ignition engine [1]. This is done by making assumptions for the evolution of the oil temperatures during warm-up and that the oil control ring during downstrokes is fully flooded. The ring-pack lubrication model includes features such as three different lubrication regimes, i.e. pure hydrodynamic lubrication, boundary lubrication and pure asperity contact, non-steady wetting of both inlet and outlet of the piston ring, capability to use all ring face profiles that can be approximated by piece-wise polynomials and, finally, the ability to model the rheology of multi-grade oils. Not surprisingly, the simulations show that by far the most important parameter is the temperature dependence of the oil viscosity. This dependence is subsequently examined further by choosing different oils. The baseline oil is SAE 10W30 and results are compared to those using the SAE 30 and the SAE 10W50 oils.

Decision tree-based context clustering based on cross validation and hierarchical priors

The standard, ad-hoc stopping criteria used in decision tree-based context clustering are known to be sub-optimal and require parameters to be tuned. This paper proposes a new approach for decision tree-based context clustering based on cross validation and hierarchical priors. Combination of cross validation and hierarchical priors within decision tree-based context clustering offers better model selection and more robust parameter estimation than conventional approaches, with no tuning parameters. Experimental results on HMM-based speech synthesis show that the proposed approach achieved significant improvements in naturalness of synthesized speech over the conventional approaches. © 2011 IEEE.

Semi-supervised video segmentation using tree structured graphical models

We present a novel, implementation friendly and occlusion aware semi-supervised video segmentation algorithm using tree structured graphical models, which delivers pixel labels alongwith their uncertainty estimates. Our motivation to employ supervision is to tackle a task-specific segmentation problem where the semantic objects are pre-defined by the user. The video model we propose for this problem is based on a tree structured approximation of a patch based undirected mixture model, which includes a novel time-series and a soft label Random Forest classifier participating in a feedback mechanism. We demonstrate the efficacy of our model in cutting out foreground objects and multi-class segmentation problems in lengthy and complex road scene sequences. Our results have wide applicability, including harvesting labelled video data for training discriminative models, shape/pose/articulation learning and large scale statistical analysis to develop priors for video segmentation. © 2011 IEEE.