10,231 research outputs found
Metadata Representations for Queryable ML Model Zoos
No full textMachine learning (ML) practitioners and organizations are building model zoos of pre-trained models, containing metadata describing properties of the ML models and datasets that are useful for reporting, auditing, reproducibility, and interpretability purposes. The metatada is currently not standardised; its expressivity is limited; and there is no interoperable way to store and query it. Consequently, model search, reuse, comparison, and composition are hindered. In this paper, we advocate for standardized ML model metadata representation and management, proposing a toolkit supported to help practitioners manage and query that metadata.Web Information SystemsHuman-Centred Artificial Intelligenc
A Manifesto of Nodalism
No full textThis paper proposes the notion of Nodalism as a means describing contemporary culture and of understanding my own creative practice in electronic music composition. It draws on theories and ideas from Kirby, Bauman, Bourriaud, Deleuze, Guatarri, and Gochenour, to demonstrate how networks of ideas or connectionist neural models of cognitive behaviour can be used to contextualize, understand and become a creative tool for the creation of contemporary electronic music
Optimizing ML Inference Queries Under Constraints
No full textThe proliferation of pre-trained ML models in public Web-based model zoos facilitates the engineering of ML pipelines to address complex inference queries over datasets and streams of unstructured content. Constructing optimal plan for a query is hard, especially when constraints (e.g. accuracy or execution time) must be taken into consideration, and the complexity of the inference query increases. To address this issue, we propose a method for optimizing ML inference queries that selects the most suitable ML models to use, as well as the order in which those models are executed. We formally define the constraint-based ML inference query optimization problem, formulate it as a Mixed Integer Programming (MIP) problem, and develop an optimizer that maximizes accuracy given constraints. This optimizer is capable of navigating a large search space to identify optimal query plans on various model zoos.Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Web Information SystemsHuman-Centred Artificial Intelligenc
Building a generalisable ML pipeline at ING
No full textAdvances in data science have caused an increase in the use of Artificial Intelligence (AI), specifically Machine Learning (ML), throughout various fields. Not only in research but in the industry as well, has ML been receiving increasing amounts of interest. Many companies rely on ML models to increase the efficiency of existing processes or offer new services and products. The industry, however, is facing several additional challenges compared to the academic context. One of those challenges is applying the Development Operations (DevOps) model to an ML application, also referred to as MLOps. This thesis sets out to find the specific challenges that practitioners encounter while operationalising ML models. To do so, we perform a single-case case study on an ML pipeline built by the Trade & Communication Surveillance team at the ING bank. This case study consists of conducting a set of interviews and performing a manual code inspection of the pipeline. The team faces challenges ranging from having insufficient time for operationalising each ML project individually to operating in the highlyregulated fintech context. Their pipeline is able to deploy a single ML model but it does not generalise well to other projects. We present the first version of an application that mitigates these challenges. The application is able to deploy ML models to the development environment at ING and can be operated by data scientists to reduce the effort of operationalising an ML model. Computer Science | Software Technolog
Hemodynamic effects of synchronous high-frequency jet ventilation during acute hypovolemia.
No full textWe studied the effects of synchronous cardiac cycle-specific high-frequency jet ventilation (HFJV) in pentobarbital-anesthetized, splenectomized, closed-chest dogs to test the hypothesis that phasic inspiratory increases in intrathoracic pressure (ITP) selectively timed to specific periods of the cardiac cycle have different hemodynamic effects during both hypovolemia (acute hemorrhage, 20 ml/kg) and neurogenic vasomotor shock (hexamethonium, 10 mg/kg) than those observed during normovolemic control conditions. Ventricular stroke volumes (SV) were measured by electromagnetic flow probes. The influence of changes in venous return (VR) on the subsequent hemodynamic response to synchronous HFJV was analyzed using instantaneous VR curves (M. R. Pinsky, J. Appl. Physiol. 56:765-771, 1984). During hemorrhage the VR curve was shifted leftward with concomitant reductions in apneic SV (15.4 +/- 3.8 to 11.2 +/- 3.6 ml, mean +/- SD), (P less than 0.01) that were accentuated by HFJV (P less than 0.01), except when the phasic inspiratory increases in ITP during HFJV were timed to occur during late diastole (-4% apneic SV, NS). SV was greater with late diastolic pulses than with other timed synchronous ITP pulses during hypovolemia (P less than 0.01). During ganglionic blockade, arterial pressure decreased (139 +/- 14 to 76 +/- 18 Torr, P less than 0.001), but VR was preserved at control levels, and no significant cardiac cycle-specific HFJV effects occurred. We conclude that SV reductions associated with positive-pressure ventilation during acute hypovolemia are minimized by HFJV synchronized to late diastole but that this effect is preload dependent
'Project smells' - Experiences in Analysing the Software Quality of ML Projects with mllint
No full textMachine Learning (ML) projects incur novel challenges in their development and productionisation over traditional software applications, though established principles and best practices in ensuring the project's software quality still apply. While using static analysis to catch code smells has been shown to improve software quality attributes, it is only a small piece of the software quality puzzle, especially in the case of ML projects given their additional challenges and lower degree of Software Engineering (SE) experience in the data scientists that develop them. We introduce the novel concept of project smells which consider deficits in project management as a more holistic perspective on software quality in ML projects. An open-source static analysis tool mllint was also implemented to help detect and mitigate these. Our research evaluates this novel concept of project smells in the industrial context of ING, a global bank and large software- and data-intensive organisation. We also investigate the perceived importance of these project smells for proof-of-concept versus production-ready ML projects, as well as the perceived obstructions and benefits to using static analysis tools such as mllint. Our findings indicate a need for context-aware static analysis tools, that fit the needs of the project at its current stage of development, while requiring minimal configuration effort from the user. Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Software EngineeringSoftware Technolog
Audiomobiles, Sculptures and Conundrums
No full textRoberto Gerhard was a pioneer of electronic music in England creating a number of substantial concert, theatre and radio works from as early as 1954. Gerhard’s electronic music is one of the richest repositories for understanding the development of the composer’s late compositional technique. Apart from the Symphony no.3, ‘Collages’, none of Gerhard’s electronic music is published. This paper will discuss aspects of Gerhard’s electronic music, focusing on Audiomobiles (1958-59) and Sculptures (1963)
Dynamic right and left ventricular interaction in tne rabbit: simultaneous measurements of ventricular pressure-volume loops.
No full textPURPOSE: This study was performed to characterize the dynamic factors determining ventricular interdependence in an open-pericardium intact animal model. MATERIALS AND METHODS: Simultaneous measures of right ventricular (RV) and left ventricular (LV) pressures and volumes in 6 urethane-anesthetized open-chested, open-pericardium rabbits. RV and LV V were calculated every 2 milliseconds. Measurements were made at initial baseline blood volume, and again after two infusions of 20 mL/kg isoconductive colloid solution. At each blood volume level, partial aortic (AO), pulmonary artery (PAO), and inferior vena caval (IVC) occlusions were performed. Biventricular diastolic compliance and end-systolic elastance were calculated from these data. RESULTS: Baseline end-diastolic (ED) and end-systolic (ES) V were 3.29 +/- 0.55 and 2.43 +/- 0.33 mL (mean +/- SD) for the LV, and 3.38 +/- 1.56 and 2.84 +/- 1.36 mL for the RV, respectively. AO increased all LV pressure and volume (P < .05) but did not alter RV ED volume (2.85 +/- 1.20 mL) or ED pressure (3.3 +/- 2.0 to 3.6 +/- 2.1 mm Hg). PAO increased RV ES pressure (P < .05) but did not alter RV ED volume, ED pressure, or ES volume, although it decreased LV ED volume (2.82 +/- 0.59, P < .05). AO also immediately increased end-systolic RV elastance to a value greater than that defined by IVC (7.9 +/- 4.4 to 10.9 +/- 6.6 mm Hg/mL, P < .05). Intravascular volume expansion though increasing baseline pressure and volume, did not alter qualitatively biventricular responses to AO, PA, or IVC. CONCLUSION: Ventricular interdependence has both systolic and diastolic components that have differing directional effects. In the pericardectomized rabbit, increases in RV ED volume decrease LV ED volume by decreasing LV diastolic compliance, but do not alter LV systolic function. Whereas, increases in LV ED volume decrease RV ES volume resulting in an increase in RV maximal elastance, but minimally alter RV diastolic function
Music for classical guitar by South African composers : a historical survey, notes on selected works and a general catalogue
No full textIncludes abstract.Includes bibliographical references (leaves 296-309).This is the first comprehensive investigation of music for, or including, the classical guitar by South African composers. The focus of this research has been, firstly, to uncover as much of the repertoire as possible, and, secondly, to collate, study, catalogue and report on the information. A brief historical survey of the guitar in South Africa provides the context within which this study was conducted. The primary sources of quantitative data collection were through the archival catalogues of the South African Music Rights Organisation and through personal contact with guitarists, composers and guitar teachers. Other sources consulted were publishers, broadcasting corporations, recording companies, libraries and the internet. The body of the dissertation comprises biographical sketches, background notes, analyses and technical notes on 17 selected solo and chamber works dating from 1947 to 2007 by some of South Africa's most prominent composers and guitaristcomposers. The repertoire ranges in style from the traditional and ethnically inspired to the experimental and abstract. As this is an empirical survey, each selected entry includes details on instrumentation, duration, level of difficulty, number of pages, scordatura, commissions or requests, sources or publishers, premières and recordings. A biography of each composer is provided as well as background notes which offer an overview of the selected work. The notes discuss historical, cultural, musical and extra-musical influences, and frequently include references to interview material. The commentaries on the selected works, with musical examples, include an analytical component describing structure, form, stylistic and compositional elements, while the technical observations include performance suggestions and a grading for each work
Cardiovascular Response in Canine Endotoxic Shock: Effect of Ibuprofen Pretreatment.
No full textIbuprofen pretreatment increases arterial pressure and reduces mortality in endotoxic dogs. The increase in arterial pressure may be caused by increases in arterial resistance, arterial sphincter tone, or both. Thus it is not clear if ibuprofen pretreatment prevents the hemodynamic effects of endotoxemia or merely masks such effects by producing concomitant increases in arterial resistance. Accordingly, this study was performed to determine the effects of ibuprofen pretreatment on arterial pressure-flow relations and other measures of cardiovascular function in a canine model of endotoxic shock. In 19 pentobarbital-anesthetized and splenectomized, closed-chest dogs, biventricular stroke volumes were measured with electromagnetic flow probes, and intrathoracic vascular and pleural pressures were measured with catheters. Instantaneous venous return curves (see Pinsky MR, J Appl Physiol 56:765, 1984) were generated during positive-pressure ventilation, and steady-state arterial pressure-flow relations, left ventricular function, peripheral vascular compliance, oxygen consumption/oxygen delivery ratio, and arterial blood lactate levels were measured during two sequential volume loading and removal (20 ml/kg) sequences. All but two dogs received a bolus infusion of Escherichia coli endotoxin (2 mg/kg) between the two fluid challenge runs. Eleven of the 17 endotoxic dogs also received ibuprofen (15 mg/kg) immediately before the initial fluid challenge. Ibuprofen pretreatment abolished all hemodynamic effects of endotoxin, whereas in the untreated group endotoxin caused decreases in calculated arterial outflow pressure and in peripheral vascular capacitance. Oxygen consumption remained constant despite changes in oxygen delivery in the nonendotoxic dogs and in the ibuprofen-pretreated dogs, whereas oxygen consumption covaried with oxygen delivery in the endotoxic dogs not pretreated with ibuprofen. Arterial lactate levels were higher after endotoxin infusion (2.1 +/- 0.5 to 3.1 +/- 0.6 mmol/liter; P less than 0.05 pre- to postvolume) but were not different between treatment groups. These data suggest that ibuprofen alters many, but not all, of the hemodynamic effects of endotoxin infusion in the dog
- …
