Discriminative training of the hidden vector state model for semantic parsing

In this paper, we discuss how discriminative training can be applied to the hidden vector state (HVS) model in different task domains. The HVS model is a discrete hidden Markov model (HMM) in which each HMM state represents the state of a push-down automaton with a finite stack size. In previous applications, maximum-likelihood estimation (MLE) is used to derive the parameters of the HVS model. However, MLE makes a number of assumptions and unfortunately some of these assumptions do not hold. Discriminative training, without making such assumptions, can improve the performance of the HVS model by discriminating the correct hypothesis from the competing hypotheses. Experiments have been conducted in two domains: the travel domain for the semantic parsing task using the DARPA Communicator data and the Air Travel Information Services (ATIS) data and the bioinformatics domain for the information extraction task using the GENIA corpus. The results demonstrate modest improvements of the performance of the HVS model using discriminative training. In the travel domain, discriminative training of the HVS model gives a relative error reduction rate of 31 percent in F-measure when compared with MLE on the DARPA Communicator data and 9 percent on the ATIS data. In the bioinformatics domain, a relative error reduction rate of 4 percent in F-measure is achieved on the GENIA corpus.

A novel framework of training hidden Markov support vector machines from lightly-annotated data

Natural language understanding (NLU) aims to map sentences to their semantic mean representations. Statistical approaches to NLU normally require fully-annotated training data where each sentence is paired with its word-level semantic annotations. In this paper, we propose a novel learning framework which trains the Hidden Markov Support Vector Machines (HM-SVMs) without the use of expensive fully-annotated data. In particular, our learning approach takes as input a training set of sentences labeled with abstract semantic annotations encoding underlying embedded structural relations and automatically induces derivation rules that map sentences to their semantic meaning representations. The proposed approach has been tested on the DARPA Communicator Data and achieved 93.18% in F-measure, which outperforms the previously proposed approaches of training the hidden vector state model or conditional random fields from unaligned data, with a relative error reduction rate of 43.3% and 10.6% being achieved.

Cleavage initiation in the intercritically reheated coarse-grained heat affected zone:Part II. Failure criteria and statistical effects

In part 1 of this article, cleavage initiation in the intercritically reheated coarse-grained heat affected zone (IC CG HAZ) of high-strength low-alloy (HSLA) steels was determined to occur between two closely spaced blocky MA particles. Blunt notch, crack tip opening displacement (CTOD), and precracked Charpy testing were used in this investigation to determine the failure criteria required for cleavage initiation to occur by this mechanism in the IC CG HAZ. It was found that the attainment of a critical level of strain was required in addition to a critical level of stress. This does not occur in the case of high strain rate testing, for example, during precracked Charpy testing. A different cleavage initiation mechanism is then found to operate. The precise fracture criteria and microstructural requirements (described in part I of this article) result in competition between potential cleavage initiation mechanisms in the IC CG HAZ.

Abnormal systemic and ocular vascular response to temperature provocation in primary open-angle glaucoma patients:a case for autonomic failure?

PURPOSE. To assess systemic and ocular vascular reactivity in response to warm and cold provocation in untreated patients with primary open-angle glaucoma and normal control subjects. METHODS. Twenty-four patients with primary open-angle glaucoma and 22 normal control subjects were subjected to a modified cold pressor test involving immersion of the right hand in 40°C warm water followed by 4°C cold water exposure, and finger and ocular blood flow were assessed by means of peripheral laser Doppler flowmetry and retinal flowmetry, respectively. Finger and body temperature as well as intraocular pressure, systemic blood pressure, systemic pulse pressure, heart rate, and ocular perfusion pressure were also monitored. RESULTS. The patients with glaucoma demonstrated an increase in diastolic blood pressure (P = 0.023), heart rate (P = 0.010), and mean ocular perfusion pressure (P = 0.039) during immersion of the tested hand in 40°C water. During cold provocation, the patients demonstrated a significant decrease in finger (P = 0.0003) and ocular blood flow (the parameter velocity measured at the temporal neuroretinal rim area; P = 0.021). Normal subjects did not demonstrate any blood flow or finger temperature changes during immersion of the tested hand in 40°C water (P > 0.05); however, they exhibited increases in systolic blood pressure (P = 0.034) and pulse pressure (P = 0.0009) and a decrease in finger blood flow (P = 0.0001) during cold provocation. In normal subjects, the ocular blood flow was unchanged during high- and low-temperature challenge. CONCLUSIONS. Cold provocation elicits a different blood pressure, and ocular blood flow response in patients with primary open-angle glaucoma compared with control subjects. These findings suggest a systemic autonomic failure and ocular vascular dysregulation in POAG patients.

Bit error rate estimation methods for QPSK CO-OFDM transmission

Coherent optical orthogonal frequency division multiplexing (CO-OFDM) is an attractive transmission technique to virtually eliminate intersymbol interference caused by chromatic dispersion and polarization-mode dispersion. Design, development, and operation of CO-OFDM systems require simple, efficient, and reliable methods of their performance evaluation. In this paper, we demonstrate an accurate bit error rate estimation method for QPSK CO-OFDM transmission based on the probability density function of the received QPSK symbols. By comparing with other known approaches, including data-aided and nondata-aided error vector magnitude, we show that the proposed method offers the most accurate estimate of the system performance for both single channel and wavelength division multiplexing QPSK CO-OFDM transmission systems. © 2014 IEEE.

Resonance vector mode locking

A mode locked fibre laser as a source of ultra-stable pulse train has revolutionised a wide range of fundamental and applied research areas by offering high peak powers, high repetition rates, femtosecond range pulse widths and a narrow linewidth. However, further progress in linewidth narrowing seems to be limited by the complexity of the carrier-envelope phase control. Here for the first time we demonstrate experimentally and theoretically a new mechanism of resonance vector self-mode locking where tuning in-cavity birefringence leads to excitation of the longitudinal modes sidebands accompanied by the resonance phase locking of sidebands with the adjacent longitudinal modes. An additional resonance with acoustic phonons provides the repetition rate tunability and linewidth narrowing down to Hz range that drastically reduces the complexity of the carrier-envelope phase control and so will open the way to advance lasers in the context of applications in metrology, spectroscopy, microwave photonics, astronomy, and telecommunications.

Phenobarbital for neonatal seizures:response rate and predictors of refractoriness

Background : Phenobarbital is the first-line choice for neonatal seizures treatment, despite a response rate of approximately 45%. Failure to respond to acute anticonvulsants is associated with poor neurodevelopmental outcome, but knowledge on predictors of refractoriness is limited. Objective : To quantify response rate to phenobarbital and to establish variables predictive of its lack of efficacy. Methods : We retrospectively evaluated newborns with electrographically confirmed neonatal seizures admitted between January 1999 and December 2012 to the neonatal intensive care unit of Parma University Hospital (Italy), excluding neonates with status epilepticus. Response was categorized as complete (cessation of clinical and electrographic seizures after phenobarbital administration), partial (reduction but not cessation of electrographic seizures with the first bolus, response to the second bolus), or absent (no response after the second bolus). Multivariate analysis was used to identify independent predictors of refractoriness. Results : Out of 91 newborns receiving phenobarbital, 57 (62.6%) responded completely, 15 (16.5%) partially, and 19 (20.9%) did not respond. Seizure type (p = 0.02), background electroencephalogram (EEG; p ≤ 0.005), and neurologic examination (p ≤ 0.005) correlated with response to phenobarbital. However, EEG (p ≤ 0.02) and seizure type (p ≤ 0.001) were the only independent predictors. Conclusion : Our results suggest a prominent role of neurophysiological variables (background EEG and electrographic-only seizure type) in predicting the absence of response to phenobarbital in high-risk newborns.

The 3 per cent Barcelona Target: A Critical Appraisal of the Failure of the Lisbon Strategy’s Knowledgebased Economy and Society Project

Ten years after the unanimous approval of the Lisbon Strategy at a special meeting of the European Council on 23-24 March 2000 in Lisbon, it will be inevitable for the European Council, the European Commission and the majority of the EU member states to face with its fi asco and to account for the reasons of their fundamental policy, governance and economic failures in 2010. The recent turbulence of the global economy offers some excuses for the underperformance of the main objectives of the Lisbon Strategy in the essential social and economic domains, like job creation, economic growth, and environmental sustainability. Negative growth rates, macroeconomic and fi nancial instability, the contraction of the internal and external markets of the European economy, drop in demand for capital investment, goods and services, sinking corporate revenues, depreciation of corporate assets, increasing private and public indebtedness, falling rate of employment, weakening social cohesion, widening social inequality, and so forth not only deprive the majority of the EU member states of fulfi lling the main objectives of the Lisbon Strategy but also drive them into worse social and economic conditions in many policy domains than they were in 2000.

Financial Failure In the Hospitality Industry

The hospitality industry (especially the restaurant segment) has a historically high rate of financial failures. Yet, financial failure in the industry has not received the attention it deserves. In this article, the authors identify basic reasons underlying failed ideas while presenting a study of several hospitality chains that have experienced varying degrees of financial failure. The characteristics and pitfalls of these companies provide the necessary groundwork to explore major lessons to be learned which should aid hospitality management to aviod future business failures.

Experimental investigation of water, snow and granular ice effects on ice failure processes and impact loads

A large series of laboratory ice crushing experiments was performed to investigate the effects of external boundary condition and indenter contact geometry on ice load magnitude under crushing conditions. Four boundary conditions were considered: dry cases, submerged cases, and cases with the presence of snow and granular ice material on the indenter surface. Indenter geometries were a flat plate, wedge shaped indenter, (reverse) conical indenter, and spherical indenter. These were impacted with artificially produced ice specimens of conical shape with 20° and 30° cone angles. All indenter – ice combinations were tested in dry and submerged environments at 1 mm/s and 100 mm/s indentation rates. Additional tests with the flat indentation plate were conducted at 10 mm/s impact velocity and a subset of scenarios with snow and granular ice material was evaluated. The tests were performed using a material testing system (MTS) machine located inside a cold room at an ambient temperature of - 7°C. Data acquisition comprised time, vertical force, and displacement. In several tests with the flat plate and wedge shaped indenter, supplementary information on local pressure patterns and contact area were obtained using tactile pressure sensors. All tests were recorded with a high speed video camera and still photos were taken before and after each test. Thin sections were taken of some specimens as well. Ice loads were found to strongly depend on contact condition, interrelated with pre-existing confinement and indentation rate. Submergence yielded higher forces, especially at the high indentation rate. This was very evident for the flat indentation plate and spherical indenter, and with restrictions for the wedge shaped indenter. No indication was found for the conical indenter. For the conical indenter it was concluded that the structural restriction due to the indenter geometry was dominating. The working surface for the water to act was not sufficient to influence the failure processes and associated ice loads. The presence of snow and granular ice significantly increased the forces at the low indentation rate (with the flat indentation plate) that were higher compared to submerged cases and far above the dry contact condition. Contact area measurements revealed a correlation of higher forces with a concurrent increase in actual contact area that depended on the respective boundary condition. In submergence, ice debris constitution was changed; ice extrusion, as well as crack development and propagation were impeded. Snow and granular ice seemed to provide additional material sources for establishing larger contact areas. The dry contact condition generally had the smallest real contact area, as well as the lowest forces. The comparison of nominal and measured contact areas revealed distinct deviations. The incorporation of those differences in contact process pressures-area relationships indicated that the overall process pressure was not substantially affected by the increased loads.

Preservation of myocardial beta-adrenergic receptor signaling delays the development of heart failure after myocardial infarction.

When the heart fails, there is often a constellation of biochemical alterations of the beta-adrenergic receptor (betaAR) signaling system, leading to the loss of cardiac inotropic reserve. betaAR down-regulation and functional uncoupling are mediated through enhanced activity of the betaAR kinase (betaARK1), the expression of which is increased in ischemic and failing myocardium. These changes are widely viewed as representing an adaptive mechanism, which protects the heart against chronic activation. In this study, we demonstrate, using in vivo intracoronary adenoviral-mediated gene delivery of a peptide inhibitor of betaARK1 (betaARKct), that the desensitization and down-regulation of betaARs seen in the failing heart may actually be maladaptive. In a rabbit model of heart failure induced by myocardial infarction, which recapitulates the biochemical betaAR abnormalities seen in human heart failure, delivery of the betaARKct transgene at the time of myocardial infarction prevents the rise in betaARK1 activity and expression and thereby maintains betaAR density and signaling at normal levels. Rather than leading to deleterious effects, cardiac function is improved, and the development of heart failure is delayed. These results appear to challenge the notion that dampening of betaAR signaling in the failing heart is protective, and they may lead to novel therapeutic strategies to treat heart disease via inhibition of betaARK1 and preservation of myocardial betaAR function.

Natriuretic peptide-based screening and collaborative care for heart failure:the STOP-HF randomized trial

IMPORTANCE: Prevention strategies for heart failure are needed.

OBJECTIVE: To determine the efficacy of a screening program using brain-type natriuretic peptide (BNP) and collaborative care in an at-risk population in reducing newly diagnosed heart failure and prevalence of significant left ventricular (LV) systolic and/or diastolic dysfunction.

DESIGN, SETTING, AND PARTICIPANTS: The St Vincent's Screening to Prevent Heart Failure Study, a parallel-group randomized trial involving 1374 participants with cardiovascular risk factors (mean age, 64.8 [SD, 10.2] years) recruited from 39 primary care practices in Ireland between January 2005 and December 2009 and followed up until December 2011 (mean follow-up, 4.2 [SD, 1.2] years).

INTERVENTION: Patients were randomly assigned to receive usual primary care (control condition; n=677) or screening with BNP testing (n=697). Intervention-group participants with BNP levels of 50 pg/mL or higher underwent echocardiography and collaborative care between their primary care physician and specialist cardiovascular service.

MAIN OUTCOMES AND MEASURES: The primary end point was prevalence of asymptomatic LV dysfunction with or without newly diagnosed heart failure. Secondary end points included emergency hospitalization for arrhythmia, transient ischemic attack, stroke, myocardial infarction, peripheral or pulmonary thrombosis/embolus, or heart failure.

RESULTS: A total of 263 patients (41.6%) in the intervention group had at least 1 BNP reading of 50 pg/mL or higher. The intervention group underwent more cardiovascular investigations (control, 496 per 1000 patient-years vs intervention, 850 per 1000 patient-years; incidence rate ratio, 1.71; 95% CI, 1.61-1.83; P<.001) and received more renin-angiotensin-aldosterone system-based therapy at follow-up (control, 49.6%; intervention, 56.5%; P=.01). The primary end point of LV dysfunction with or without heart failure was met in 59 (8.7%) of 677 in the control group and 37 (5.3%) of 697 in the intervention group (odds ratio [OR], 0.55; 95% CI, 0.37-0.82; P = .003). Asymptomatic LV dysfunction was found in 45 (6.6%) of 677 control-group patients and 30 (4.3%) of 697 intervention-group patients (OR, 0.57; 95% CI, 0.37-0.88; P = .01). Heart failure occurred in 14 (2.1%) of 677 control-group patients and 7 (1.0%) of 697 intervention-group patients (OR, 0.48; 95% CI, 0.20-1.20; P = .12). The incidence rates of emergency hospitalization for major cardiovascular events were 40.4 per 1000 patient-years in the control group vs 22.3 per 1000 patient-years in the intervention group (incidence rate ratio, 0.60; 95% CI, 0.45-0.81; P = .002).

CONCLUSION AND RELEVANCE: Among patients at risk of heart failure, BNP-based screening and collaborative care reduced the combined rates of LV systolic dysfunction, diastolic dysfunction, and heart failure.

TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00921960.

Biomarker Testing and Pre-emptive Therapy in Preventing Heart Failure

In an attempt to reduce the heart failure epidemic,screening and prevention will become an increasing focus ofmanagement in the wider at-risk population. Refining riskprediction through the use of biomarkers in isolation or incombination is emerging as a critical step in this process.The utility of biomarkers to identify disease manifestationsbefore the onset of symptoms and detrimental myocardialdamage is proving to be valuable. In addition, biomarkers thatpredict the likelihood and rate of disease progression over timewill help streamline and focus clinical efforts and therapeuticstrategies. Importantly, several recent early intervention studiesusing biomarker strategies are promising and indicate thatnot only can new-onset heart failure be reduced but also thedevelopment of other cardiovascular conditions.

Predictive Analytics and Decision Support for Heart Failure patients

Thesis (Master's)--University of Washington, 2016-08

Prediction of Sudden Cardiac Arrest for Patients with Congestive Heart Failure

Thesis (Ph.D.)--University of Washington, 2016-07