Sökning: "semantic representation"

Visar resultat 6 - 10 av 76 uppsatser innehållade orden semantic representation.

  1. 6. Multi-Scale Task Dynamics in Transfer and Multi-Task Learning : Towards Efficient Perception for Autonomous Driving

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

    Författare :Simon Ekman von Huth; [2023]
    Nyckelord :Autonomous Driving; Computer Vision; Deep Learning; Machine Learning; Multi-Task Learning; Transfer Learning; Task Relationships; Task Dynamics; Python; Multi-Scale Representation Learning; Fuss-Free Network; Självkörande Fordon; Datorseende; Djupinlärning; Maskininlärning; Multiuppgiftsinlärning; Överföringsinlärning; Uppgiftsrelationer; Uppgiftsdynamik; Python; Flerskalig Representationsinlärning; Fuss-Free Nätverk;

    Sammanfattning : Autonomous driving technology has the potential to revolutionize the way we think about transportation and its impact on society. Perceiving the environment is a key aspect of autonomous driving, which involves multiple computer vision tasks. LÄS MER

  3. 7. Digital Twin Knowledge Graphs for IoT Platforms : Towards a Virtual Model for Real-Time Knowledge Representation in IoT Platforms

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

    Författare :Alejandro Jarabo Peñas; [2023]
    Nyckelord :Internet of Things; Digital Twin; Knowledge Graph; Similarity Metric; Semantic Data Integration; Internet of Things; Digital Twin; Kunskapsgraf; Likhetmetrik; Integrering av Semantiska Data; Internet of Things; Gemelo Digital; Grafo de Conocimiento; Métrica de Similitud; Integración de Datos Semánticos.;

    Sammanfattning : This thesis presents the design and prototype implementation of a digital twin based on a knowledge graph for Internet of Things (IoT) platforms. The digital twin is a virtual representation of a physical object or system that must continually integrate and update knowledge in rapidly changing environments. LÄS MER

  4. 8. The role of word accents in semantic processing in South Swedish

    Master-uppsats, Lunds universitet/Masterprogram: Språk och språkvetenskap; Lunds universitet/Fonetik

    Författare :Jinhee Kwon; [2023]
    Nyckelord :Word accents; spoken word recognition; prediction; ERP; N400; Languages and Literatures;

    Sammanfattning : Prosodic cues can aid speech processing by adding semantic information in lexical tones or functional information in intonational tones. Swedish word accents are considered to have stronger grammatical functions than semantic roles, although they are shaped by both lexical and intonational information structure. LÄS MER

  5. 9. Using a Deep Generative Model to Generate and Manipulate 3D Object Representation

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

    Författare :Yu Hu; [2023]
    Nyckelord :Neural networks; point cloud; 3D shape generation; 3D shape manipulation; classification; Neurala nätverk; punktmoln; generering av 3D-former; manipulation av 3Dformer; klassificering;

    Sammanfattning : The increasing importance of 3D data in various domains, such as computer vision, robotics, medical analysis, augmented reality, and virtual reality, has gained giant research interest in generating 3D data using deep generative models. The challenging problem is how to build generative models to synthesize diverse and realistic 3D objects representations, while having controllability for manipulating the shape attributes of 3D objects. LÄS MER

  7. 10. Messing With The Gap: On The Modality Gap Phenomenon In Multimodal Contrastive Representation Learning

    Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Industriell teknik

    Författare :Mohammad Al-Jaff; [2023]
    Nyckelord :Multimodal machine learning; Representation learning; Self-supervised learning; contrastive learning; computer vision; computational biology; bioinformatics;

    Sammanfattning : In machine learning, a sub-field of computer science, a two-tower architecture model is a specialised type of neural network model that encodes paired data from different modalities (like text and images, sound and video, or proteomics and gene expression profiles) into a shared latent representation space. However, when training these models using a specific contrastive loss function, known as the multimodalinfoNCE loss, seems to often lead to a unique geometric phenomenon known as the modality gap. LÄS MER