114612 research outputs found
Sort by
Teaching-track economists in Canada, the United Kingdom, and the United States
No full textFor Canada, the United Kingdom, and the United States, we illuminate the landscape for a relatively new and evolving role: full-time, teaching-track economists who work in the same departments as research-track economists, but with a greater emphasis on te aching. We use in-depth interviews and a large-scale survey. We employ a mixed-methods approach. A cohesive, cross-country, multi-institution comparison enables learning from a variety of contexts. Our findings inform decision-making processes, initiate co nversations among multiple constituents, generate ideas, raise salient questions, and identify relative strengths and weaknesses of different teaching-track models
Machine unlearning in learned databases : an experimental analysis
No full textMachine learning models based on neural networks (NNs) are enjoying ever-increasing attention in the Database (DB) community, both in research and practice. However, an important issue has been largely overlooked, namely the challenge of dealing with the inherent, highly dynamic nature of DBs, where data updates are fundamental, highly-frequent operations (unlike, for instance, in ML classification tasks). Although some recent research has addressed the issues of maintaining updated NN models in the presence of new data insertions, the effects of data deletions (a.k.a., "machine unlearning") remain a blind spot. With this work, for the first time to our knowledge, we pose and answer the following key questions: What is the effect of unlearning algorithms on NN-based DB models? How do these effects translate to effects on key downstream DB tasks, such as cardinality/selectivity estimation (SE), approximate query processing (AQP), data generation (DG), and upstream tasks like data classification (DC)? What metrics should we use to assess the impact and efficacy of unlearning algorithms in learned DBs? Is the problem of (and solutions for) machine unlearning in DBs different from that of machine learning in DBs in the face of data insertions? Is the problem of (and solutions for) machine unlearning for DBs different from unlearning in the ML literature? what are the overhead and efficiency of unlearning algorithms (versus the naive solution of retraining from scratch)? What is the sensitivity of unlearning on batching delete operations (in order to reduce model updating overheads)? If we have a suitable unlearning algorithm (forgetting old knowledge), can we combine it with an algorithm handling data insertions (new knowledge) en route to solving the general adaptability/updatability requirement in learned DBs in the face of both data inserts and deletes? We answer these questions using a comprehensive set of experiments, various unlearning algorithms, a variety of downstream DB tasks (such as SE, AQP, and DG), and an upstream task (DC), each with different NNs, and using a variety of metrics (model-internal, and downstream-task specific) on a variety of real datasets, making this also a first key step towards a benchmark for learned DB unlearning
Polydimethylsiloxane (PDMS) coated broadband unable vanadium dioxide (VO2) based linear optical cavity temperature sensor
No full textSilicon on insulator (SOI) based sensors provide a reasonable solution to the issues common in traditional linear optical cavities such as wavelength dependant nature of mirrors, size, and maintaining the resonant condition. In this study we presented polydimethylsiloxane (PDMS) coated SOI based linear optical temperature sensing resonator model and analysed it in finite element method by using COMSOL Multiphysics. The phase changing material (PCM) VO2 on each side of the Si waveguide helped to achieve the resonant condition and thermal tunability of the resonator. An almost linear variation in resonant frequency (wavelength) fr due to the temperature change in the range of 0–90 °C resulted in maximum sensitivity of 0.01 THz/°C or 79.4 pm/°C for the 10 µm cavity length. The recorded sensitivity is at least 5-times (or more) higher than the previous studies. The prominent reasons behind this improvement can be PDMS coating, adequate light matter interaction and proper confinement of resonating mode. The demonstrated sensor model has wide operational frequency range spanning from 10 to 210 THz. Moreover, the reported model also showed an increase in temperature sensitivity from 0.00967 to 0.01 THz/°C while the length of resonator was changed from 2 to 10 µm
Measurement of the Centrality Dependence of the Dijet Yield in
No full textATLAS measured the centrality dependence of the dijet yield using 165 nb−1 o
Meniscal transplant surgery or optimised rehabilitation full randomised trial (MeTeOR2) : a study protocol
No full textIntroduction: Pain and disability after meniscectomy can be a substantial lifelong problem. There are few treatment options, especially for young people. Non-surgical management (rehabilitation) is an option but increasingly surgeons are performing meniscal allograft transplants (MATs) for these individuals. However, this is still an uncommon procedure, and availability and usage of MAT vary widely both in the UK and internationally. It is not known which treatment option is the most effective and cost-effective.
Methods and analysis: The Meniscal Transplant surgery or Optimised Rehabilitation trial is an international, multicentre, randomised controlled trial. The aim is to compare the clinical and cost effectiveness of MAT versus an optimised package of individualised, progressive, rehabilitation that we have called personalised knee therapy (PKT).Participants will be recruited from sites across the UK, Australia, Canada and Belgium. The planned 144 participants provide at least 90% power to detect a 10-point difference in the Knee injury and Osteoarthritis Outcome Score (KOOS4) at 24-months post randomisation (primary outcome). A prospectively planned economic evaluation will be conducted from a healthcare system and personal social services perspective. Secondary outcome data including health utility, occupational status, sports participation, mental well-being, further treatment, and adverse events will be collected at 3, 6, 12, 18, and 24 months. Analysis will be on an intention-to-treat basis and reported in-line with the Consolidated Standards of Reporting Trials statement.
Ethics and dissemination: The trial was approved by the London-Bloomsbury Research Ethics Committee on 19 August 2022 (22/LO/0327) and Northern Sydney Local Health District Human Research Ethics Committee, NSW, Australia on the 13 March 2023 (2022/ETH01890).Trial results will be disseminated via peer-reviewed publications, presentations at international conferences, in lay summaries and using social media as appropriate. This protocol adheres to the recommended Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist.
Trial registration number: ISRCTN87336549.
Keywords: Knee; REHABILITATION MEDICINE; Randomized Controlled Trial; SURGERY; TRANSPLANT SURGERY
Supporting patients with a mental health diagnosis to use online services in primary care. A qualitative interview study
No full textObjective: The increase in reliance on online services for general practice has the potential to increase inequalities within some populations. Patients with a mental health condition are one such group. Digital facilitation is defined as a range of processes, procedures, and people, which seek to support NHS patients in using online services. This study aimed to examine the views and experiences of digital facilitation in primary care amongst patients living with a mental health condition. Methods: Semi-structured interviews were conducted with patients living with a mental health condition, recruited from general practices across England participating in the Di-Facto study. Thematic analysis was conducted on interview transcripts. Results: Interviews were conducted with ten participants with a mental health condition, recruited from five general practices. Three themes were identified: (1) familiarity with online services; (2) experiences of those using online services; (3) the need for digital facilitation. The need for digital facilitation was identified in the registration for online services, and in trusting online services. Conclusions: Online services offer convenience for patients, but registration for the use of such services remains a potential area of difficulty. Participants had difficulties with registering for online services and had concerns about trust in using them. Support offered by general practices in using online services needs to be varied and adaptable to meet the needs of individual patients
Flipping sensemaking on its head : from common sense to sensus communis
No full textSensemaking provides a compelling account of how meaning emerges by theorizing the organizational enactment of order. In this paper we question the underlying assumption that making sense is equivalent to ordering. We draw from Hannah Arendt’s work to argue that restricting sense to ordering as a means of addressing practical concerns is limiting, and even dehumanizing, and that the most profound forms of sense may emerge from disrupting rather than restoring order. In questioning the intimacy between sense and order, we also question the common-sense view that organization seeks practical settlements, certainty and reliability. Following Arendt, we pursue the question of what it means to organize for plural opinion-making, a condition she conceptualizes as sensus communis. The upshot is to flip sensemaking on its head: Rather than meaning being generated through organizing, and certain types of disruption merely triggering it, sense is made through disruption, with certain types of organizing enabling it
