Model Selection for Unsupervised Learning of Visual Context

This study addresses the problem of choosing the most suitable probabilistic model selection criterion for unsupervised learning of visual context of a dynamic scene using mixture models. A rectified Bayesian Information Criterion (BICr) and a Completed Likelihood Akaike’s Information Criterion (CL-AIC) are formulated to estimate the optimal model order (complexity) for a given visual scene. Both criteria are designed to overcome poor model selection by existing popular criteria when the data sample size varies from small to large and the true mixture distribution kernel functions differ from the assumed ones. Extensive experiments on learning visual context for dynamic scene modelling are carried out to demonstrate the effectiveness of BICr and CL-AIC, compared to that of existing popular model selection criteria including BIC, AIC and Integrated Completed Likelihood (ICL). Our study suggests that for learning visual context using a mixture model, BICr is the most appropriate criterion given sparse data, while CL-AIC should be chosen given moderate or large data sample sizes.     

PETROLEUM MIGRATION, FAULTS AND OVERPRESSURE, PART I: CALIBRATING BASIN MODELLING USING PETROLEUM IN TRAPS — A REVIEW

This paper considers the principles of deciphering basin-scale hydrocarbon migration patterns using the geochemical information which is present in trapped petroleum. Petroleum accumulations in subsiding basins can be thought of as "data archives" within which stored information can help us to understand aspects of hydrocarbon formation and migration. This information can impart a time-resolved picture of hydrocarbon migration in a basin in response to processes associated with progressive burial, particularly in the context of the occurrence and periodic activity of faults.
This review, which includes a series of tentative models of migration-related processes in the extensional Halten Terrace area, offshore mid-Norway, illustrates how we can use information from the migrating mobile hydrocarbon phase to improve our knowledge of the static geological system. Of particular importance is the role of sub-seismic heterogeneities and faults in controlling migration processes. We focus on how the secondary migration process can be enhanced in a multi-source rock basin such as the Halten Terrace, thereby increasing prospectivity.     

A Novel Approach to Analyzing the Occupancy of the Geo/Geo/∞ Queueing System

Discrete time queueing systems have been successfully used to model packet queues in computer and digital communications systems. In this article we introduce a novel approach to analyzing the Geo^X/Geo^Y/∞ queueing system, which is suitable for modelling high speed computer and communications systems. The approach is based on characterizing the system by a functional equation which is then solved recursively to give identical results to those obtained using other approaches.     

Simulated recruitment of riparian trees and shrubs under natural and regulated flow regimes on the Wisconsin River,USA

Models that link ecological responses to hydrologic changes are important for assessing the effects of flow regulation on aquatic and riparian ecosystems. Based on the Recruitment Box Model, a graphical model used to prescribe environmental flows for cottonwood (Populus spp.) recruitment, we designed a simulation model to represent the influence of river flow dynamics on seedling recruitment of riparian pioneer woody plants. The model simulates the influence of temporal patterns of river stage on dispersal, germination, initial recruitment and over‐winter survival of first‐year seedlings of riparian pioneer shrubs and trees. We used the model to simulate seedling recruitment patterns for five species (Acer saccharinum, Betula nigra, Populus deltoides, Salix nigra and Salix exigua) on the Wisconsin River (Wisconsin, USA) under three flow scenarios: historic (1935–2002), simulated natural (1915–1975) and simulated regulated flows (1915–1975). Simulation results agreed well with field‐observed relative differences among years (1997–2000) in seedling densities for the five focal species. Simulated successful recruitment years were highly synchronous among species, but species differed in their sensitivity to flows at different times during the growing season, consistent with among‐species differences in seed dispersal timing. Comparison of simulated natural and regulated flows for 1915–1975 showed that flow regulation decreased monthly flow variability, increased late summer to winter baseflow and reduced the magnitude of spring peaks. Simulated recruitment and over‐winter survival of tree seedlings of all species was enhanced under the regulated flow scenario, likely due to increased summer baseflow and reductions in peak flood magnitude. Our analyses show the utility of extending the Recruitment Box Model to include multiple species of riparian shrubs and trees, and the effects of post‐colonization flows on their recruitment success. However, some key functional relationships between flow patterns and woody seedling demography (e.g. shear stress thresholds for seedling mortality) have not been adequately quantified and merit further study. Copyright © 2006 John Wiley & Sons, Ltd.     

A concept for the generation of out-of-plane distortion from tailored FRP laminates

The phenomenon of thermally induced distortions in unsymmetrical laminates is well understood, and it may be shown that a square, unsymmetrical 0.90 laminate will tend to form two stable geometries with a snap-through phenomenon between them. This paper discusses laminates in which at each point the lay-up is symmetrical across the laminate mid-plane, but which still exhibit multiple stable geometries. The number of stable geometries can be controlled by the details of the lay-up from the minimum of two to, in principle, an unlimited number. In addition it will be shown how a similar process can be used to generate multiple stable stress states and geometries in unidirectional laminates. This paper represents a very preliminary experimental investigation of the design space available for such composite laminates. Possible applications of composites with multiple stable geometries are noted.     

机器人腕力传感器的动态负载效应

本文从时域和频域两方面研究机器人腕力传感器在加载情况下的动态响应，进行动态建模和频谱分析，探讨负载效应的影响。     

可移动机器人的运动学模型与控制原理

本文首先对可移动机器人的运动学进行了分析，建立了运动学模型与控制结构图，然后采用时域分析法对此系统进行了校正，最后通过对此系统的仿真与时测实验，对各校正环节的参数进行了整定。     

Materials modelling: A bridge from atoms to bulk properties

The use of molecular-level materials modelling techniques in the development of advanced performance polymers is discussed, with particular emphasis upon bridging the large difference in the scales of dimensions between atomic structure and fabricated parts. The advantages and disadvantages of bulk quantitative structure-property relations and atomistic modelling are assessed, and the method of group interaction modelling is suggested as a means of bridging the dimensional scales.After a brief introduction to the concept of group interaction modelling, examples of modelling the engineering properties of polymers are presented which are difficult to model quantitatively by any other means. The important phase transitions from the crystal and glassy states of matter to those of rubber- and liquidlike states are shown quantitatively to be due to the same isoenergetic condition. The viscoelastic properties of a polymer are critical for many applications and expressions are derived for the loss and storage components of the complex modulus, with reference to failure initiation conditions. The effect of crosslinking in thermosets upon the glass transition temperature and viscoelastic properties is outlined. Finally, the scaling of time from atomic vibrations to the years involved in creep and ageing effects are discussed.     

Adaptive user modelling in intelligent telephone and email assistants

With the burgeoning complexity and capabilities of modern information appliances and services, user modelling is becoming an increasingly important research area. Simple user profiles already personalise many software products and consumer goods such as digital TV recorders and mobile phones. A user model should be easy to initialise, and it must adapt in the light of interaction with the user. In many cases, a large amount of training data is needed to generate a user model, and adaptation is equivalent to retraining the system. This paper briefly outlines the user modelling problem and work done at BTexact on an intelligent personal assistant (IPA) which incorporates a user profile. We go on to describe FILUM, a more flexible method of user modelling, and show its application to the telephone assistant and email assistant components of the IPA, with tests to illustrate its usefulness.