9,732 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
'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)
Gut induced biomarkers of appetite and satiety
No full textObesity is basically caused by an imbalance between food intake and energy expenditure. It is well documented that the gastrointestinal tract plays a key role in the control of food intake, but the regulatory circuits and their interactions are complex.
Food enters the gastrointestinal tract, which then trigger specific mechanisms that respond to specific components of food. The anatomical bases for the sensing machinery are enteroendocrine cells in the small intestine, which act as neural triggers or as intestinal satiation peptide secreting cells. These cells express chemosensory receptors that respond to luminal stimuli. This thesis addresses specific mechanisms regarding enteroendocrine cells and how nutrient components interact with this machinery to stimulate and regulate the secretion of gut peptides, which play a key role in the regulation of food intake and a wide range of metabolic functions.
In a first set of experiments, we investigated the involvement of two potential targets of peptide release, such as glucagon-like peptide 1 (GLP-1), peptide tyrosine tyrosine (PYY) and cholecystokinin (CCK): i) bile acids (BAs) as possible TGR5 agonists and ii) glucose stimulating the sweet receptor T1R2/T1R3.
To investigate the physiological role of BAs, subjects received intraduodenal infusions of different loads of chenodeoxycholic acid (CDCA, a primary BA in humans) in comparison to sodium-oleate (a potent secretagogue for the peptides mentioned above) or vehicle as a control. Administration of CDCA resulted in a significant increase of both plasma GLP-1 and CCK levels; however, the stimulatory potency was small, if we compare the magnitude of the GLP-1 and CCK responses to other well-known secretagogues such as glucose or fatty acids.
To investigate the physiological role of T1R2/T1R3 in the secretion of intestinal satiation peptides we used lactisole, a T1R2/T1R3 receptor antagonist. Subjects received i) intragastric and intraduodenal infusions of glucose and ii) intragastric and intraduodenal infusions of a liquid mixed meal, both with and without lactisole. Lactisole induced a significant reduction of plasma GLP-1 levels in both, the intragastric and intraduodenal glucose-stimulated parts. However, we observed no effect of lactisole on gastrointestinal peptide secretion in the mixed liquid meal-stimulated parts. The liquid meal consisted beside glucose also of proteins, fats and other complex carbohydrates. The lack of effect of lactisole suggests that these nutrients induced the release of gastrointestinal peptides probably via other receptor mechanisms and thus outweighed the effect of T1R2/T1R3 blockade. These findings indicate that the receptor is not alone responsible for peptide secretion; it is rather a complex interaction between different receptor mechanisms. In addition, we found that the inhibitory effect of lactisole on the secretion of GLP-1 was greater in response to intragastric glucose administration compared to the intraduodenal infusion. These results let assume interaction mechanisms between gastric signals and signals from the small intestine and indicate a relevant contribution of the stomach in the regulation of gastrointestinal peptide secretion.
Indeed, several studies in animals and humans suggest that gastric and intestinal signals interact to elicit optimal satiation and adequate control of eating. In humans, little information is available on the underlying mechanisms of this interaction. In addition, uncertainties exist about the role of both gastric and intestinal parameters, as well as their interaction in the control of satiation in relation to body mass. In a second set of experience, we investigated the reciprocal control between gastric functions and intestinal parameters in the control of appetite in lean as well as in obese persons.
To investigate this potential interaction, lean subjects received either a rapid intragastric load or a continuous intraduodenal infusion of glucose or a mixed liquid meal. We found that infusions of glucose directly into the small intestine elicit only weak effects on appetite and the secretion of GLP-1 and PYY. In contrast, identical amounts of glucose delivered into the stomach markedly suppressed appetite paralleled by significantly greater plasma levels of GLP-1 and PYY. Administration of the mixed liquid meal showed a similar outcome. It seems that an initial more rapid rate of duodenal delivery after intragastric infusions account for the accelerated secretion of GLP-1 and PYY. These findings suggest again a role of the stomach in the control of appetite and indicate interaction mechanisms between gastric emptying rates and the release of intestinal satiation peptides.
In a last series of experiments, we compared gastric emptying, intestinal peptide release and satiation parameters in response to nutrients between normal weight and obese healthy subjects. We found that gastric emptying rates were delayed in obese subjects, possible due to impaired gastric sensory functions. In addition, the increase in post-prandial plasma GLP-1 and PYY levels was reduced and the caloric intake was higher in obese compared to lean subjects. These results document once more the importance of gastric signals in the control of appetite.
Together, chemosensing receptors like T1R2/T1R3 are involved in the secretion of gastrointestinal peptides, however each receptor by itself is probably not alone responsible for peptide release – it is rather a complex interaction between different receptor mechanisms. In addition, complex interactions between different gastrointestinal signals are responsible for the control of eating. The understanding of each of these signals and interaction mechanisms is essential and could constitute a promising therapeutic approach for the treatment of obesity
- …
