Sökning: "Distribution inbäddningar"

Hittade 3 uppsatser innehållade orden Distribution inbäddningar.

  1. 1. Optimizing the Performance of Text Classification Models by Improving the Isotropy of the Embeddings using a Joint Loss Function

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

    Författare :Joseph Attieh; [2022]
    Nyckelord :Text Classification; Isotropy; Embeddings; BERT; IsoScore; Klassificering av Text; Isotropi; Inbäddningar; BERT; IsoScore;

    Sammanfattning : Recent studies show that the spatial distribution of the sentence representations generated from pre-trained language models is highly anisotropic, meaning that the representations are not uniformly distributed among the directions of the embedding space. Thus, the expressiveness of the embedding space is limited, as the embeddings are less distinguishable and less diverse. LÄS MER

  3. 2. Classification of Transcribed Voice Recordings : Determining the Claim Type of Recordings Submitted by Swedish Insurance Clients

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

    Författare :Carl Piehl; [2021]
    Nyckelord :Text Classification; Word embeddings; BERT; LSTM; Cost-sensitive learning; Focal loss; Textklassificering; Ordinbäddningar; BERT; LSTM; Kostnadskänslig inlärning; Fokal förlustfunktion;

    Sammanfattning : In this thesis, we investigate the problem of building a text classifier for transcribed voice recordings submitted by insurance clients. We compare different models in the context of two tasks. The first is a binary classification problem, where the models are tasked with determining if a transcript belongs to a particular type or not. LÄS MER

  4. 3. Improving Zero-Shot Learning via Distribution Embeddings

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

    Författare :Vivek Chalumuri; [2020]
    Nyckelord :Zero-Shot Learning ZSL ; Generalized Zero-Shot Learning GZSL ; Image Classification; Metric Learning; Distribution Embeddings; Triplet Loss; Zero-shot lärande; Generaliserat zero-shot-lärande; Bildklassificering; Metrisk inlärning; Distribution inbäddningar; Triplet-förlust;

    Sammanfattning : Zero-Shot Learning (ZSL) for image classification aims to recognize images from novel classes for which we have no training examples. A common approach to tackling such a problem is by transferring knowledge from seen to unseen classes using some auxiliary semantic information of class labels in the form of class embeddings. LÄS MER