Sökning: "model confidence set"

Visar resultat 1 - 5 av 72 uppsatser innehållade orden model confidence set.

  1. 1. To Deploy, or Not to Deploy, That is the Question A qualitative study of the decision-making experiences of engineers deploying software changes in production

    Magister-uppsats, Lunds universitet/Avdelningen för Riskhantering och Samhällssäkerhet

    Författare :Jessica Ann DeVita; [2024]
    Nyckelord :Automation; Common ground; Complexity; Coordination; COVID19 pandemic; Deployment decision; Incident response; Mental model; Naturalistic decision-making; Phenomenology; Resilience; Risk; Software engineering; SRE; Strategy; Temporal reasoning; Uncertainty; Web operations; FLMU06; Technology and Engineering;

    Sammanfattning : The web services that millions of people count on require large-scale computer resources (servers, nodes, networking) and an intricate set of interdependent software services. Under pressure to continually improve these services, code and configuration changes are deployed hundreds or thousands of times every day. LÄS MER

  3. 2. Fine-tuning a BERT-based NER Model for Positive Energy Districts

    Master-uppsats, Högskolan Dalarna/Institutionen för information och teknik

    Författare :Karen Ortega; Fei Sun; [2023]
    Nyckelord :Positive Energy District PED ; Named Entity Recognition NER ; Bidirectional Encoder Representations from Transformers BERT ; Pipeline; Fine-tune;

    Sammanfattning : This research presents an innovative approach to extracting information from Positive Energy Districts (PEDs), urban areas generating surplus energy. PEDs are integral to the European Commission's SET Plan, tackling housing challenges arising from population growth. LÄS MER

  4. 3. Bootstrapping methods for assessing causality in survival analysis: A case study on major adverse cardiovascular events

    Master-uppsats, Lunds universitet/Matematisk statistik

    Författare :Paulina Benthem Ciano; [2023]
    Nyckelord :Causality; Graphical models; Bootstrap; Survival analysis; Additive hazards model.; Mathematics and Statistics;

    Sammanfattning : The combination of graphical models with Aalen's additive hazards model, resulting in a model known as dynamical path analysis, permits assessing the effects of different variables on times until an event and decomposing these total effects into direct and indirect effects. This thesis proposes novel bootstrapping methods designed for left-truncated right-censored data, conditional on covariates within the framework of Aalen's additive hazards model, in order to obtain confidence intervals for the estimates. LÄS MER

  5. 4. Constraining the Higgs width in Higgs production associated with a top quark pair

    Kandidat-uppsats, Lunds universitet/Fysiska institutionen; Lunds universitet/Teoretisk partikelfysik - Geonomgår omorganisation

    Författare :Benjamin Dahlén; [2023]
    Nyckelord :Higgs; width; cross section; on-shell; off-shell; invariant mass; Standard model; upper bound; Physics and Astronomy;

    Sammanfattning : In this thesis we study the width of the Higgs boson in the process pp → t ̄t 4l at LO using MadGraph5_aMC@NLO to generate the events. The contributions including the Higgs signal, continuum background and interference were considered in order to calculate the expected number of events in a broad range of four-lepton invariant masses. LÄS MER

  7. 5. The effect of model calibration on noisy label detection

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Max Joel Söderberg; [2023]
    Nyckelord :Area Under the Margin; Calibration; Image classification; Label smoothing; Noisy labels; Overconfidence; Area under marginalen; Kalibrering; Bildklassificering; Etikettutjämning; Brusiga etiketter; Övertro;

    Sammanfattning : The advances in deep neural networks in recent years have opened up the possibility of using image classification as a valuable tool in various areas, such as medical diagnosis from x-ray images. However, training deep neural networks requires large amounts of annotated data which has to be labelled manually, by a person. LÄS MER