9,789 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
Qualità dell’aria e concentrazioni elementari in capelli e urine di studenti universitari italiani e cileni
No full textIl monitoraggio di matrici biologiche umane (capelli, urine) può consentire di ottenere informazioni sull’esposizione inalatoria tramite una misura diretta delle concentrazioni di sostanze chimiche nelle diverse matrici (indicatori di esposizione e di effetto). Le concentrazioni elementari costituiscono uno tra
gli indicatori di esposizione più diffusi ed affidabili. Nel presente studio sono state valutate le concentrazioni di 41 elementi nei capelli e nelle urine di studenti di due università: una situata a Roma e una a Santiago del Cile. Nei due campus universitari, parallelamente ai campionamenti biologici, sono stati raccolti campioni di PM10 mediante l’uso di campionatori a basso flusso (0.5 L min-1), sui quali sono state effettuate analisi del contenuto elementare e di potenziale ossidativo (OP; misurato mediante i saggi acellulari del DCFH, AA e DTT). Tutte le analisi elementari sono state condotte mediante spettroscopia atomica a plasma accoppiato induttivamente (ICPMS) e spettroscopia di fluorescenza atomica (HG-AFS; 1,2). Sui dati ottenuti è stata valutata l’influenza di alcune variabili relative agli studenti partecipanti quali l’età, il genere, l’indice di massa corporea, il fumo e la dieta.
La concentrazione media di PM10 è risultata pari a 26 μg m-3 a Roma e a 71 μg m-3 a Santiago del Cile, indicando livelli di esposizione inalatoria molto differenti nelle due aree. In particolare, i livelli di Cu, Mo, Sn e Sb, traccianti del traffico veicolare non combustivo (abrasione meccanica dei freni; 3,4),
mostrano concentrazioni nettamente più elevate nei campioni di PM10 cileni rispetto a quelli italiani. Anche i valori di OP misurati mediante i saggi del DCFH e del AA sono risultati nettamente più elevati al sito cileno, confermando un contributo rilevante del traffico veicolare non combustivo al potenziale ossidativo delle polveri. Le diverse concentrazioni elementari nel PM10 si sono tradotte in diversi livelli di accumulo di metalli e metalloidi nelle due matrici biologiche studiate. Gli elementi che hanno presentato differenze maggiormente significative tra gli studenti italiani e cileni in entrambe le matrici biologiche sono stati Cs, Li, Mo, Sb e Sr. Al contrario, Cu, Al e Sn variano significativamente sia nelle urine che nei capelli non in riferimento all’area geografica ma in dipendenza del genere dei partecipanti. Il possibile incremento dei livelli di stress ossidativo nelle urine dei soggetti esposti all’aria di Santiago del Cile verrà verificato mediante l’analisi di marcatori di effetto, come l’oxo-guanina e altri prodotti di ossidazione del DNA ed RNA.
Bibliografia
[1] Astolfi ML et al. 2020. DOI: 10.3390/ijerph17061911
[2] Astolfi ML et al. 2020. DOI: 10.1039/C9AY01871A
[3] Thorpe A and Harrison RM 2008. DOI: 10.1016/j.scitotenv.2008.06.007
[4] Massimi L et al. 2020. DOI: 10.1016/j.envpol.2020.11527
Effects of terbuthylazine on iron-deficient barley: interferences on phytosiderophores release and on sulfur metabolism
No full textEffetto della terbutilazina in piante di orzo ferro carenti: rilascio di fitosiderofori e metabolismo dello zolfo
No full textOptimizing 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
Effects of terbuthylazine on phytosiderophores release in iron deficient barley
No full textOne of the main purposes of modern agriculture is to raise crop production by means of a proper
application of herbicides. Nonetheless, the use of herbicides is causing some concerns associated to their
persistence and accumulation in the environment. Notwithstanding the relevance of these aspects, little
or no attention has been paid to the interferences that these chemicals and their residues can exert on
iron (Fe) mineral nutrition of plants. To this purpose, we studied the effect of terbuthylazine (TBA) on the
Fe-acquisition processes of Fe-deficient barley plants, by investigating some aspects related to
phytosiderophores (PS) exudation and sulfur (S) metabolism. Results showed that plant growth,
chlorophyll content expressed as SPAD index, PS release and the expression of genes involved in their
secretion, uptake and biosynthesis were negatively affected. This response was associated with reduced
cysteine concentrations, and it suggests that the TBA interferences on Fe-acquisition in barley are the
consequence of the induced changes in S metabolism
A SCADA-Based Method for Estimating the Energy Improvement from Wind Turbine Retrofitting
No full textRetrofitting of operating wind turbines has been spreading in the recent years in the wind energy industry, at the aim of optimizing the efficiency of wind kinetic energy conversion. This kind of interventions has material and labor costs and it is therefore fundamental to estimate realistically the production improvement. Further, the retrofitting of wind turbines sited in harsh environments, as for example complex terrain, might exacerbate the stressing conditions and therefore affect the residue lifetime. This work deals with a case of retrofitting: the testing ground is a multi-megawatt wind turbine from a wind farm sited in a very complex terrain. The blades have been optimized by installing vortex generators and passive flow control devices. A general method is proposed for estimating production upgrades from wind turbine retrofitting, basing on multivariate linear modeling of the power output of the upgraded wind turbine. Applying the model to the test case of interest, it arises that the upgrade increases the annual energy production of the wind turbine of an amount of the order of the 2%
Impairment of the Zn/Cd detoxification systems affects the ability of Salmonella to colonize Arabidopsis thaliana
No full textSalmonella capacity to colonize different environments depends on its ability to respond efficiently to fluctuations in micronutrient availability. Among micronutrients, Zn, besides playing an essential role in bacterial physiology, is a key element whose concentration can influence bacterial survival in a particular niche. Plant colonization by Salmonella enterica was described for several years, and some molecular determinants involved in this host-pathogen interaction have started to be characterized. However, it is still unclear if Zn plays a role in the outcome of this interaction, as well established for animal hosts that employ nutritional immunity strategies to counteract pathogens infections. In this study, we have investigated the involvement of Salmonella Typhimurium main effectors of zinc homeostasis in plant colonization, using Arabidopsis thaliana as a model host. The results show that to colonize plant tissues, Salmonella takes advantage of its ability to export excess metal through the efflux pumps ZntA and ZitB. In fact, the deletion of these Zn/Cd detoxification systems can affect bacterial persistence in the shoots, depending on metal availability in the plant tissues. The importance of Salmonella ability to export excess metal was enhanced in the colonization of plants grown in high Zn conditions. On the contrary, the bacterial disadvantage related to Zn detoxification impairment can be abrogated if the plant cannot efficiently translocate Zn to the shoots. Overall, our work highlights the role of Zn in Salmonella-plant interaction and suggests that modulation of plant metal content through biofortification may be an efficient strategy to control pathogen colonization
- …
