Sökning: "encoder"

Visar resultat 16 - 20 av 264 uppsatser innehållade ordet encoder.

1. 16. Anomaly Detection of Time Series Caused by International Revenue Share Fraud : Additive Model and Autoencoder Applications

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

   Författare :Lingxiao Wang; [2023]
   Nyckelord :Fraud detection; Anomaly detection; Machine learning; Bedrägeriupptäckt; Anomalidetektering; Maskininlärning;

   Sammanfattning : In this paper, we compare the performance of two methods to find the attempts at fraud from the data provided by Sinch (formerly CLX Communications, which is a telecommunications and cloud communications platform as a service (PaaS) company). We consider the problem as finding the anomaly in a time series signal, where we ignore the duration of a single call or other features and only care about the total volume of calls in a certain period. LÄS MER

3. 17. Regularizing Vision-Transformers Using Gumbel-Softmax Distributions on Echocardiography Data

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

   Författare :Alfred Nilsson; [2023]
   Nyckelord :Deep Learning; Vision-Transformers; Echocardiography; Feature Selection; Gumbel-Softmax; Concrete Autoencoders; Regression; Djupinlärning; Vision-Transformers; Ekokardiografi; Feature Selection; GumbelSoftmax; Concrete Autoencoders; Regression;

   Sammanfattning : This thesis introduces an novel approach to model regularization in Vision Transformers (ViTs), a category of deep learning models. It employs stochastic embedded feature selection within the context of echocardiography video analysis, specifically focusing on the EchoNet-Dynamic dataset. LÄS MER

4. 18. A Transformer-Based Scoring Approach for Startup Success Prediction : Utilizing Deep Learning Architectures and Multivariate Time Series Classification to Predict Successful Companies

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

   Författare :Gustaf Halvardsson; [2023]
   Nyckelord :Machine learning; Time Series Classification; Transformers; Gated Recurrent Unit; Venture Capital; Maskininlärning; tidsseriesklassifiering; Transformer; Gated Recurrent Unit; riskkapital;

   Sammanfattning : The Transformer, an attention-based deep learning architecture, has shown promising capabilities in both Natural Language Processing and Computer Vision. Recently, it has also been applied to time series classification, which has traditionally used statistical methods or the Gated Recurrent Unit (GRU). LÄS MER

5. 19. Designing a Cost Effective Encoder for High Performance Doors

   Uppsats för yrkesexamina på avancerad nivå, Lunds universitet/Industriell elektroteknik och automation

   Författare :Emanuel Edelstam; David Petersson; [2023]
   Nyckelord :Encoder; High speed doors; Motor; Product development; Prototyping; ASSA ABLOY; Technology and Engineering;

   Sammanfattning : Målet med detta arbete var att utveckla en prototyp av en kostnadseffektiv pulsgivare till motorerna för ASSA ABLOYs snabbrullduksportar. Pulsgivaren ger återkoppling till de automatiska dörrarnas kontrollsystem och ska göra detta på ett tillförlitligt sätt samtidigt som de ska vara mer kostnadseffektiva än de nuvarande lösningarna som används på ASSA ABLOY. LÄS MER

7. 20. Fault Detection and Diagnosis for Automotive Camera using Unsupervised Learning

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

   Författare :Ziyou Li; [2023]
   Nyckelord :Unsupervised Learning; Autoencoders; Image Clustering; Fault Detection and Diagnosis; Morphological Operations; Hardware-in-Loop; Advanced DriverAssistance System; Oövervakad inlärning; Autoencoders; Bildklustering; Felfindning och Diagnostik; Morfologiska Operationer; Hardware-in-Loop; Avancerade Förarassistanssystem;

   Sammanfattning : This thesis aims to investigate a fault detection and diagnosis system for automotive cameras using unsupervised learning. 1) Can a front-looking wide-angle camera image dataset be created using Hardware-in-Loop (HIL) simulations? 2) Can an Adversarial Autoencoder (AAE) based unsupervised camera fault detection and diagnosis method be crafted for SPA2 Vehicle Control Unit (VCU) using an image dataset created using Hardware-inLoop? 3) Does using AAE surpass the performance of using Variational Autoencoder (VAE) for the unsupervised automotive camera fault diagnosis model? In the field of camera fault studies, automotive cameras stand out for its complex operational context, particularly in Advanced Driver-Assistance Systems (ADAS) applications. LÄS MER