Stochastic neural networks

The first purpose of this paper is to present a class of algorithms for finding the global minimum of a continuous-variable function defined on a hypercube. These algorithms, based on both diffusion processes and simulated annealing, are implementable as analog integrated circuits. Such circuits can be viewed as generalizations of neural networks of the Hopfield type, and are called diffusion machines.Our second objective is to show that learning in these networks can be achieved by a set of three interconnected diffusion machines: one that learns, one to model the desired behavior, and one to compute the weight changes.This research was supported in part by U.S. Army Research Office Grant DAAL03-89-K-0128.     

A COMPARATIVE STUDY OF FEATURE SELECTION FOR HIDDEN MARKOV MODEL-BASED MICRO-MILLING TOOL WEAR MONITORING

This paper presents a discriminant feature selection approach for hidden Markov model (HMM) modeling of micro-milling tool conditions. The approach is compared with other popular feature selection methods such as principal component analysis (PCA) and automatic relevance determination (ARD) according to their HMM classification rate. In tool condition monitoring (TCM), there are a lot of features that contain redundant information or less sensitive to tool state discrimination. These features are expected to be deleted for less computation and more robust modeling of tool conditions. Fisher's linear discriminant analysis (FDA) is modified for this purpose. The FDA is generally used for classification, and the features are mapped to another space and lose their physical meanings. In the modified discriminant feature selection, the features are selected in the original feature space by maximizing tool state separation and ranked by their separation ability between different tool states. Experimental results from both micro-milling of copper and steel under different working conditions indicate that the FDA is superior to both PCA and ARD for feature selection in HMM's classification. The reasons behind these differences are also discussed.     

Source Choice Perception and Sustainable Rural Water Supply Development: A Case Study of Ban Thadindam,Lopburi, Thailand

Abstract

This study seeks to examine source choice perception and sustainable rural water supply development in Ban Thadindam, a Thai village situated about 220 km north of Bangkok. It makes use of an elementary schematic matrix that was initially developed for the valuation of water sources as perceived by water users in East Africa. In applying the elementary choice scheme, the author interviewed the village Kamnan (district officer) and a few villagers. The results of this study show that a wide range of water supply sources was perceived by the Kamnan. Out of a practical range of six sources, four were perceived and accepted, roof (rain water), river, spring, and pond; and two were perceived but not accepted: well and canal. The reason why well and canal water sources were rejected was because of their hardness and turbidity. Of the four sources accepted, roof (rain water) was rated very favorably as the preferred choice while river, spring, and pond sources were judged equally favorable. Awareness of source choice and the conditions under which choice is made is crucial for the understanding and development of sustainable rural water supply. By adopting a wide combination of sources and integrating them to provide a rich water supply for treatment, and by taking advantage of filtration technology given by the Asian Institute of Technology, Environmental Engineering Division, the villagers, with the help of the Project Managing Committee, transformed the demonstration filtration facility into a sustainable rural water supply system. The latter succeeded not only in providing adequate filtered water for the majority of the households in Ban Thadindam but also in satisfying the water requirements for the whole village for two decades. The success of the rural water supply system has been attributable to many factors, but most important was the awareness the villagers have in the understanding of the interrelations between landscape, range of choice and the interaction between community participation, technology, and environment.     

A dark-line two-dimensional magneto-optical trap of (85)Rb atoms with high optical depth

We describe the apparatus of a dark-line two-dimensional (2D) magneto-optical trap (MOT) of (85)Rb cold atoms with high optical depth (OD). Different from the conventional configuration, two (of three) pairs of trapping laser beams in our 2D MOT setup do not follow the symmetry axes of the quadrupole magnetic field: they are aligned with 45° angles to the longitudinal axis. Two orthogonal repumping laser beams have a dark-line volume in the longitudinal axis at their cross over. With a total trapping laser power of 40 mW and repumping laser power of 18 mW, we obtain an atomic OD up to 160 in an electromagnetically induced transparency (EIT) scheme, which corresponds to an atomic-density-length product NL = 2.05 × 10(15) m(-2). In a closed two-state system, the OD can become as large as more than 600. Our 2D MOT configuration allows full optical access of the atoms in its longitudinal direction without interfering with the trapping and repumping laser beams spatially. Moreover, the zero magnetic field along the longitudinal axis allows the cold atoms maintain a long ground-state coherence time without switching off the MOT magnetic field, which makes it possible to operate the MOT at a high repetition rate and a high duty cycle. Our 2D MOT is ideal for atomic-ensemble-based quantum optics applications, such as EIT, entangled photon pair generation, optical quantum memory, and quantum information processing.     

Dynamical synapses enhance neural information processing: gracefulness, accuracy, and mobility

Experimental data have revealed that neuronal connection efficacy exhibits two forms of short-term plasticity: short-term depression (STD) and short-term facilitation (STF). They have time constants residing between fast neural signaling and rapid learning and may serve as substrates for neural systems manipulating temporal information on relevant timescales. This study investigates the impact of STD and STF on the dynamics of continuous attractor neural networks and their potential roles in neural information processing. We find that STD endows the network with slow-decaying plateau behaviors: the network that is initially being stimulated to an active state decays to a silent state very slowly on the timescale of STD rather than on that of neuralsignaling. This provides a mechanism for neural systems to hold sensory memory easily and shut off persistent activities gracefully. With STF, we find that the network can hold a memory trace of external inputs in the facilitated neuronal interactions, which provides a way to stabilize the network response to noisy inputs, leading to improved accuracy in population decoding. Furthermore, we find that STD increases the mobility of the network states. The increased mobility enhances the tracking performance of the network in response to time-varying stimuli, leading to anticipative neural responses. In general, we find that STD and STP tend to have opposite effects on network dynamics and complementary computational advantages, suggesting that the brain may employ a strategy of weighting them differentially depending on the computational purpose.     

Recognizing and measuring self-regulated learning in a mobile learning environment

With the realization that more research is needed to explore external factors (e.g., pedagogy, parental involvement in the context of K-12 learning) and internal factors (e.g., prior knowledge, motivation) underlying student-centered mobile learning, the present study conceptually and empirically explores how the theories and methodologies of self-regulated learning (SRL) can help us analyze and understand the processes of mobile learning. The empirical data collected from two elementary science classes in Singapore indicates that the analytical SRL model of mobile learning proposed in this study can illuminate the relationships between three aspects of mobile learning: students’ self-reports of psychological processes, patterns of online learning behavior in the mobile learning environment (MLE), and learning achievement. Statistical analyses produce three main findings. First, student motivation in this case can account for whether and to what degree the students can actively engage in mobile learning activities metacognitively, motivationally, and behaviorally. Second, the effect of students’ self-reported motivation on their learning achievement is mediated by their behavioral engagement in a pre-designed activity in the MLE. Third, students’ perception of parental autonomy support is not only associated with their motivation in school learning, but also associated with their actual behaviors in self-regulating their learning.     

Joint segmentation of collectively moving objects using a bag-of-words model and level set evolution

In scenes with collectively moving objects, to disregard the individual objects and take the entire group into consideration for motion characterization is a promising approach with wide application prospects. In contrast to studies on the segmentation of independently moving objects, our purpose is to construct a segmentation of these objects to characterize their motions at a macroscopic level. In general, the collectively moving objects in a group have very similar motion behavior with their neighbors and appear as a kind of global collective motion. This paper presents a joint segmentation approach for these collectively moving objects. In our model, we extract these macroscopic movement patterns based on optical flow field sequences. Specifically, a group of collectively moving objects correspond to a region where the optical flow field has high magnitude and high local direction coherence. As a result, our problem can be addressed by identifying these coherent optical flow field regions. The segmentation is performed through the minimization of a variational energy functional derived from the Bayes classification rule. Specifically, we use a bag-of-words model to generate a codebook as a collection of prototypical optical flow patterns, and the class-conditional probability density functions for different regions are determined based on these patterns. Finally, the minimization of our proposed energy functional results in the gradient descent evolution of segmentation boundaries which are implicitly represented through level sets. The application of our proposed approach is to segment and track multiple groups of collectively moving objects in a large variety of real-world scenes.     

Generalized Adjusted Rand Indices for cluster ensembles

In this paper, Adjusted Rand Index (ARI) is generalized to two new measures based on matrix comparison: (i) Adjusted Rand Index between a similarity matrix and a cluster partition (ARImp), to evaluate the consistency of a set of clustering solutions with their corresponding consensus matrix in a cluster ensemble, and (ii) Adjusted Rand Index between similarity matrices (ARImm), to evaluate the consistency between two similarity matrices. Desirable properties of ARI are preserved in the two new measures, and new properties are discussed. These properties include: (i) detection of uncorrelatedness; (ii) computation of ARImp/ARImm in a distributed environment; and (iii) characterization of the degree of uncertainty of a consensus matrix. All of these properties are investigated from both the perspectives of theoretical analysis and experimental validation. We have also performed a number of experiments to show the usefulness and effectiveness of the two proposed measures in practical applications.     

Service-oriented architecture for ontologies supporting multi-agent system negotiations in virtual enterprise

This paper offers a service-oriented architecture (SOA) for ontology-based multi-agent system (MAS) negotiations in the context of virtual enterprises (VEs). The objective of this paper is fourfold. First, it is to design a SOA which utilizes ontology and MAS to provide a distributed and interoperable environment for automated negotiations in VE. In this architecture, individual ontologies for both the VE initiator and its potential partners are constructed to describe and store resources and service knowledge. Second, a series of semantic ontology matching methods are developed to reach agents’ interoperability during the negotiation process. Third, correspondence-based extended contract net protocol is presented, which provides basic guidelines for agents’ reaching mutual understandings and service negotiation. Last, a fuzzy set theory based knowledge reuse approach is proposed to evaluate the current negotiation behaviors of the VE partners. A walkthrough example is presented to illustrate the methodologies and system architecture proposed in this paper.