Sökning: "model averaging"

Visar resultat 1 - 5 av 67 uppsatser innehållade orden model averaging.

1. 1. Traffic State Estimation on Swedish Highways : Model Comparison using Multisource Data

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

   Författare :Jiaqi Xu; [2023]
   Nyckelord :Traffic State Estimation; Macroscopic Traffic Model; Extended Kalman Filter; Particle Filter; Data Fusion; Trafiklägesuppskattning; Makroskopisk trafikmodell; Utökad Kalman-filter; Partikelfilter; Datafusion;

   Sammanfattning : Due to the escalating demand for traffic information and management, the significance of traffic state estimation, which involves the assessment of traffic conditions on road segments with limited measurement data, is increasing. Two primary estimation methods are model-driven and data-driven. LÄS MER

3. 2. On Linear Mode Connectivity up to Permutation of Hidden Neurons in Neural Network : When does Weight Averaging work?

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

   Författare :Adhithyan Kalaivanan; [2023]
   Nyckelord :Mode Connectivity; Representation Learning; Loss Landscape; Network Symmetry; Lägesanslutning; representationsinlärning; förlustlandskap; nätverkssymmetri;

   Sammanfattning : Neural networks trained using gradient-based optimization methods exhibit a surprising phenomenon known as mode connectivity, where two independently trained network weights are not isolated low loss minima in the parameter space. Instead, they can be connected by simple curves along which the loss remains low. LÄS MER

4. 3. Complex KPIs versus the usual benchmarks : a case study on Svensk Dos order picking department

   Kandidat-uppsats, SLU/Dept. of Economics

   Författare :Shahin Armaki; Kawa Ahmad Mohammed; [2023]
   Nyckelord :Warehousing; Regression; Analysis; Sweden; Pharmaceutical Industry;

   Sammanfattning : This thesis addresses the research gap of a limited discussion on how to establish targets and expectations for Key Performance Indicators (KPIs) related to order processing in warehouse settings. It emphasizes the scarcity of studies addressing this problem and the challenges faced by larger companies in monitoring their order processing performance without clear benchmarks. LÄS MER

5. 4. Deep Learning-based Regularizers for Cone Beam Computed Tomography Reconstruction

   Master-uppsats, KTH/Matematisk statistik

   Författare :Sabina Syed; Josefin Stenberg; [2023]
   Nyckelord :Adversarial Convex Regularization; Computer Vision; Cone Beam Computed Tomography; Convolutional Neural Networks; Deep Learning; Image Reconstruction; Adversarial Convex Regularization; Bildrekonstruktion; Datorseende; Djupinlärning; Faltningsnätverk; Volymtomografi;

   Sammanfattning : Cone Beam Computed Tomography is a technology to visualize the 3D interior anatomy of a patient. It is important for image-guided radiation therapy in cancer treatment. During a scan, iterative methods are often used for the image reconstruction step. LÄS MER

7. 5. Evaluation of FMCW Radar Jamming Sensitivity

   Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Signaler och system

   Författare :Ludvig Snihs; [2023]
   Nyckelord :FMCW; FMCW radar; radar; frequency-modulated radar; noise jamming; frequency-modulated continuous wave radar; FMCW Jamming; Jamming; Pulse Train; Pulse Jamming; Spoofing Attack; Spoofing; Deception; Repeater Jamming; Repeater; Automotive Radar; CFAR; sensor jamming; constant false alarm rate; electronic warfare; EW; FMCW; FMCW radar; frekvensmodulerad dopplerradar; dopplerradar; radar; brusstörning; radarstörning; pulsstörning; pulståg; falskmål; vilseledning; repeterstörning; bilradar; sensorstörning; CFAR;

   Sammanfattning : In this work, the interference sensitivity of an FMCW radar has been evaluated by studying the impact on a simulated detection chain. A commercially available FMCW radar was first characterized and its properties then laid the foundation for a simulation model implemented in Matlab. LÄS MER