10,642 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
Spontaneous music : the first generation British free improvisers
No full textThe British free improvisation scene originated in London and Sheffield during the
mid 1960s. In groups such as AMM, the Spontaneous Music Ensemble and Joseph
Holbrooke, a distinctive and ambitious musicality developed that still occupies most
of its protagonists forty years later.
Marked stylistic contrasts developed within the genre, notably the `atomistic' and
`laminar' methods of interaction. Nonetheless, a consistency of principle and practice
was also apparent that defined British free improvisation as unique. In some respects
the genre resembled its German, Dutch and American counterparts, and also the jazz
and classical avant-gardes that had inspired them. Both conceptually and practically,
however, clear differences remained.
The British free improvisers refined a method and an aesthetic of musical creativity,
which suggested an intimate perspective and a detailed analysis of that which we
accept as `music'. Its techniques and results were unconventional, but remained
consistent with music's defining concepts and experiences. As such, British free
improvisation suggested a more inclusive model of musicality than is common, and
implied a broad critique of the cultural values that define `music' at all. Though the
free improvisers themselves did not explicitly state the connection, their work may be
viewed in the context of Deconstruction: the post-structuralist analytical strategy
associated with philosopher Jacques Derrida.
British free improvisation culminated from innovations within the twentieth century
avant-garde. Referencing styles such as atonality and free jazz, it challenged the
aesthetic, technical and hierarchical standards of Western tradition in a form that was
striking and extreme, but also of logical development and focus. Free improvisation
owed explicit debt to a variety of other musics; its most singular achievement
however, was the redefinition of `rhythm' by which it disguised this fact.
The music of the first generation British free improvisers is reliant upon precise
conceptual and practical execution. But though this has enabled the genre to be
musically innovative, in the long term it has also become a logical problem. With
British free improvisation as its subject, the scrutiny of Deconstruction reveals
significant discrepancies between what `free improvisation' implies and what it
actually represents
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)
A well-conserved Plasmodium falciparum var gene shows an unusual stage-specific transcript pattern
No full textThe var multicopy gene family encodes Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) variant antigens, which, through their ability to adhere to a variety of host receptors, are thought to be important virulence factors. The predominant expression of a single cytoadherent PfEMP1 type on an infected red blood cell, and the switching between different PfEMP1 types to evade host protective antibody responses, are processes thought to be controlled at the transcriptional level. Contradictory data have been published on the timing of var gene transcription. Reverse transcription-polymerase chain reaction (RT-PCR) data suggested that transcription of the predominant var gene occurs in the later (pigmented trophozoite) stages, whereas Northern blot data indicated such transcripts only in early (ring) stages. We investigated this discrepancy by Northern blot, with probes covering a diverse var gene repertoire. We confirm that almost all var transcript types were detected only in ring stages. However, one type, the well-conserved varCSA transcript, was present constitutively in different laboratory parasites and does not appear to undergo antigenic variation. Although varCSA has been shown to encode a chondroitin sulphate A (CSA)-binding PfEMP1, we find that the presence of full-length varCSA transcripts does not correlate with the CSA-binding phenotype
A longitudinal study of motor, oculomotor and cognitive function in progressive supranuclear palsy
No full textObjectiveWe studied the annual change in measures of motor, oculomotor and cognitive function in progressive supranuclear palsy. This had twin objectives, to assess the potential for clinical parameters to monitor disease progression in clinical trials and to illuminate the progression of pathophysiology.MethodsTwenty three patients with progressive supranuclear palsy (Richardson’s syndrome) were compared to 22 matched controls at baseline and 16 of these patients compared at baseline and one year using: the progressive supranuclear palsy rating scale; the unified Parkinson’s disease rating scale; the revised Addenbrooke’s cognitive examination; the frontal assessment battery; the cubes section of the visual object and space perception battery; the Hayling and Brixton executive tests; and saccadic latencies.ResultsPatients were significantly impaired in all domains at baseline. However, cognitive performance was maintained over a year on the majority of tests. The unified Parkinson’s disease rating scale, saccadic latency and progressive supranuclear palsy rating scale deteriorated over a year, with the latter showing the largest change. Power estimates indicate that using the progressive supranuclear palsy rating scale as an outcome measure in a clinical trial would require 45 patients per arm, to identify a 50% reduction in rate of decline with 80% power.ConclusionsMotor, oculomotor and cognitive domains deteriorate at different rates in progressive supranuclear palsy. This may be due to differential degeneration of their respective cortical-subcortical circuits, and has major implications for the selection of outcome measures in clinical trials due to wide variation in sensitivity to annual rates of decline
Discovery of a second SALMFamide gene in the sea urchin Strongylocentrotus purpuratus reveals that L-type and F-type SALMFamide neuropeptides coexist in an echinoderm species
No full textNOTICE: this is the author’s version of a work that was accepted for publication in MARINE GENOMICS. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in MARINE GENOMICS, [VOL 3, ISSUE 2, (2010)] DOI: 10.1016/j.margen.2010.08.00
- …
