FASTBUS Demonstration Systems

This paper will provide a demonstration of basic FASTBUS hardware and test software. The systems will include single crate segments, simple computer I/O, a fast sequencer and memory, some simple diagnostic and display devices and a UNIBUS to FASTBUS processor interface. The equipment will be set up to show the basic FASTBUS protocols and timing transactions, as well as some of the general initialization software features.     

A segmented topic model based on the two-parameter Poisson-Dirichlet process

Documents come naturally with structure: a section contains paragraphs which itself contains sentences; a blog page contains a sequence of comments and links to related blogs. Structure, of course, implies something about shared topics. In this paper we take the simplest form of structure, a document consisting of multiple segments, as the basis for a new form of topic model. To make this computationally feasible, and to allow the form of collapsed Gibbs sampling that has worked well to date with topic models, we use the marginalized posterior of a two-parameter Poisson-Dirichlet process (or Pitman-Yor process) to handle the hierarchical modelling. Experiments using either paragraphs or sentences as segments show the method significantly outperforms standard topic models on either whole document or segment, and previous segmented models, based on the held-out perplexity measure.     

Unsupervised Object Discovery: A Comparison

The goal of this paper is to evaluate and compare models and methods for learning to recognize basic entities in images in an unsupervised setting. In other words, we want to discover the objects present in the images by analyzing unlabeled data and searching for re-occurring patterns. We experiment with various baseline methods, methods based on latent variable models, as well as spectral clustering methods. The results are presented and compared both on subsets of Caltech256 and MSRC2, data sets that are larger and more challenging and that include more object classes than what has previously been reported in the literature. A rigorous framework for evaluating unsupervised object discovery methods is proposed.     

Efficient parameter learning of Bayesian network classifiers

Recent advances have demonstrated substantial benefits from learning with both generative and discriminative parameters. On the one hand, generative approaches address the estimation of the parameters of the joint distribution—\(\mathrm{P}(y,\mathbf{x})\), which for most network types is very computationally efficient (a notable exception to this are Markov networks) and on the other hand, discriminative approaches address the estimation of the parameters of the posterior distribution—and, are more effective for classification, since they fit \(\mathrm{P}(y|\mathbf{x})\) directly. However, discriminative approaches are less computationally efficient as the normalization factor in the conditional log-likelihood precludes the derivation of closed-form estimation of parameters. This paper introduces a new discriminative parameter learning method for Bayesian network classifiers that combines in an elegant fashion parameters learned using both generative and discriminative methods. The proposed method is discriminative in nature, but uses estimates of generative probabilities to speed-up the optimization process. A second contribution is to propose a simple framework to characterize the parameter learning task for Bayesian network classifiers. We conduct an extensive set of experiments on 72 standard datasets and demonstrate that our proposed discriminative parameterization provides an efficient alternative to other state-of-the-art parameterizations.     

11—AN INVESTIGATION OF THE AIR-JET BULKING PROCESS

Experiments aimed at investigating the mechanism of the air-jet bulking or texturing process led to an interesting observation that the textured yarn emerging from the venturi was rotating in such a way as would temporarily untwist the yarn while it was in the jet. The indication is that false-twisting (or untwisting) occurs owing to the vorticity of the turbulent air in the wake of the yarn feed needle that bifurcates the air-stream. The various yarn-processing conditions separately affect the amount of false twist inserted. There is evidence to show that the direction of the vortex, and hence of the untwisting action, is dependent on whether the angular displacement of the feed needle from its central position is in a clockwise or anti-clockwise direction. A suggested mechanism of the process based on the untwisting and retwisting of the overfed filaments offers an explanation for the characteristic locked-in loop structure of air-bulked yarns.     

29—CAM-TO-NEEDLE IMPACT FORCES IN WEFT-KNITTING PART VI: A GUARD-CAM IMPACT TRANSDUCER

The design of a transducer for measuring the impact force existing between the needle and the guard cam at the instant of contact is outlined. A typical trial result is given.     

24—CAM-TO-NEEDLE IMPACT FORCES IN WEFT-KNITTING PART I: THE THEORY OF STITCH-CAM IMPACT

Before a transducer could be designed to measure the reaction force between the stitch cam and a needle at the instant of contact, it was necessary to estimate the pulse shape, frequency, and magnitude of the force. An analytical method is here described for determining the nature of the force pulse.     

Alcohol-related infractions among college students: Associations with subsequent drinking as a function of sensitivity to punishment.

Problematic alcohol use on college campuses is a significant concern. Violations of campus alcohol policies can lead to disciplinary action from the university. These and other alcohol-related legal infractions may be a sign of significant alcohol-related problems. However, few studies have focused on determining predictors of alcohol-related infractions among college students. Likewise, the role of infractions in reducing future use is unclear. In the present study, we tested whether alcohol-related infractions were associated with decreased alcohol use, and whether the effect of the infraction varied as a function of initial drinking levels, sensitivity to punishment (SP), and sensitivity to reward (SR) in a 6-month prospective design. Alcohol use, grade point average, and SR were significantly associated with receiving an alcohol-related infraction. For heavier drinkers, receiving an infraction was associated with decreased drinking at follow-up, and this decrease was most pronounced among those with higher sensitivity to punishment. SP appeared to increase responsiveness to the infraction, resulting in greater attenuation of drinking at follow-up. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

32—AN ANALYSIS OF THE AIR-JET YARN-TEXTURING PROCESS PART VII: THE EFFECTS OF PROCESSING PARAMETERS ON YARN PROPERTIES

The properties of the supply yarn and the processing parameters together determine the final properties of air-jet-textured yarns. An investigation is reported on how variations in overfeed, air-pressure, and texturing speed, wet or dry processing, and the use of an impact element each affect the properties of the yarn textured. The test plan and the test methods are described, prior to a report of the test results. It is concluded that the test results agree with the predictions made from the postulated mechanism of loop formation.