11,553 research outputs found

    Metadata Representations for Queryable ML Model Zoos

    No full text
    Machine 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 text
    This 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 text
    The 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

    A pharmacometric model describing the relationship between warfarin dose and INR response with respect to variations in CYP2C9, VKORC1, and age

    No full text
    The objective of the study was to update a previous NONMEM model to describe the relationship between warfarin dose and international normalized ratio (INR) response, to decrease the dependence of the model on pharmacokinetic (PK) data, and to improve the characterization of rare genotype combinations. The effects of age and CYP2C9 genotype on S-warfarin clearance were estimated from high-quality PK data. Thereafter, a temporal dose-response (K-PD) model was developed from information on dose, INR, age, and CYP2C9 and VKORC1 genotype, with drug clearance as a covariate. Two transit compartment chains accounted for the delay between exposure and response. CYP2C9 genotype was identified as the single most important predictor of required dose, causing a difference of up to 4.2-fold in the maintenance dose. VKORC1 accounted for a difference of up to 2.1-fold in dose, and age reduced the dose requirement by similar to 6% per decade. This reformulated K-PD model decreases dependence on PK data and enables robust assessment of INR response and dose predictions, even in individuals with rare genotype combinations

    A PK-PD model for predicting the impact of age, CYP2C9, and VKORC1 genotype on individualization of warfarin therapy

    No full text
    The aim of this study was to characterize the relationship between warfarin concentrations and international normalized ratio (INR) response and to identify predictors important for dose individualization. S- and R-warfarin concentrations, INR, and CYP2C9 and VKORC1 genotypes from 150 patients were used to develop a population pharmacokinetic/ pharmacodynamic model in NONMEM. The anticoagulant response was best described by an inhibitory EMAX model, with S-warfarin concentration as the only exposure predictor for response. Delay between exposure and response was accounted for by a transit compartment model with two parallel transit compartment chains. CYP2C9 genotype and age were identified as predictors for S-warfarin clearance, and VKORC1 genotype as a predictor for warfarin sensitivity. Predicted INR curves indicate important steady-state differences between patients with different sets of covariates; differences that cannot be foreseen from early INR assessments alone. It is important to account for CYP2C9 and VKORC1 genotypes and age to improve a priori and a posteriori individualization of warfarin therapy

    Continuous Subcutaneous Administration of Miso Extracts Attenuates Salt-Induced Hypertension in Dahl Salt-Sensitive Rats

    No full text
    Objective: Long-term intake of Miso attenuates hypertension in Dahl salt-sensitive (Dahl S) rats through an increased urinary sodium excretion. We examined whether a bolus injection into the peritoneal cavity (i.p.) or a continuous subcutaneous infusion of a Miso extract attenuates hyper-tension in Dahl S rats. Materials and Methods: We investigated the effects of a bolus, i.p. injection of 50 mg Miso extract in 0.5 mL on hypertension in Dahl S rats, and examined whether a long-term subcutaneous infusion of the Miso extract (50 mg Miso/day), using an osmotic mini-pump working for 14 days, attenuates hypertension in Dahl S rats. Results: A bolus, i.p. injection of 50 mg Miso extract decreased SBP in 2 hrs. The reduction was significant at 4 hrs, and SBP returned to the baseline at 24 hrs. The SBP reduction at 4 hrs after the injection was increasingly greater each day during the 4 days. The SBP reduction by the Miso extract was dose-dependent and the antihy-pertensive activity occurs in a <5 kDa fraction of the extract. The subcutaneous infusion of 50 mg Miso extract/day for 14 days significantly attenuated hypertension in Dahl S rats. The SBP reduc-tion was associated with significant decreases in the heart and kidney weights. Urinary protei
    corecore