Essays about: "gradient descent"

Showing result 1 - 5 of 69 essays containing the words gradient descent.

  1. 1. Variational AutoEncoders and Differential Privacy : balancing data synthesis and privacy constraints

    University essay from KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Author : Baptiste Bremond; [2024]
    Keywords : TVAE; Differential privacy; Tabular data; Synthetic data; DP-SGD; TVAE; differentiell integritet; tabelldata; syntetiska data; DP-SGD;

    Abstract : This thesis investigates the effectiveness of Tabular Variational Auto Encoders (TVAEs) in generating high-quality synthetic tabular data and assesses their compliance with differential privacy principles. The study shows that while TVAEs are better than VAEs at generating synthetic data that faithfully reproduces the distribution of real data as measured by the Synthetic Data Vault (SDV) metrics, the latter does not guarantee that the synthetic data is up to the task in practical industrial applications. READ MORE

  2. 2. Control perspective on distributed optimization

    University essay from Uppsala universitet/Sannolikhetsteori och kombinatorik

    Author : Sam Farkhooi; [2023]
    Keywords : Distributed optimization; control; PI; gradient descent;

    Abstract : In the intersection between machine learning, artificial intelligence and mathe- matical computation lies optimization. A powerful tool that enables us to solve a variety of large scale problems. The purpose of this work is to explore optimiza- tion in the distributed setting. READ MORE

  3. 3. Multiple-Emitter Super-resolution Imaging using the Alternating Descent Conditional Gradient Method

    University essay from Lunds universitet/Beräkningsbiologi och biologisk fysik - Genomgår omorganisation; Lunds universitet/Institutionen för astronomi och teoretisk fysik - Genomgår omorganisation

    Author : Dolev Illouz; [2023]
    Keywords : super-resolution; alternating descent conditional gradient method; ADCG; multiple-emitter localization; nanoscale imaging; single molecule tracking; Physics and Astronomy;

    Abstract : This thesis examines the state-of-the-art 2D super-resolution technique alternating descent conditional gradient (ADCG) method's ability to accurately localize fluorophores in diffraction-limited single molecule images (SMI) and analyze the impact of pre-processing and post-processing modules on ADCG's fluorophore localization. A synthetic dataset obtained from the 2013 Grand Challenge localization microscopy and a temporally linked dataset obtained from an unpublished set of Optical DNA mapping experiments performed by Jonathan Jeffet at the NanoBioPhotonix Lab at Tel-Aviv University were initially segmented to extract their noise parameters. READ MORE

  4. 4. Exploration of Alternative Image Representation Using Signed Distance Functions

    University essay from Lunds universitet/Institutionen för designvetenskaper

    Author : Jakub Olejnik; [2023]
    Keywords : computer graphics; SDF-based rendering; machine learning; Technology and Engineering;

    Abstract : This project leverages the core concepts of differentiable rendering to find alternative representation of images using geometric primitives rendered with Signed Distance Functions. By utilising gradient-descent-based methods it is possible to find optimal parameters of rendered primitives to create high-quality reconstruction. READ MORE

  5. 5. Methods for determining focal point and delay for ultrasound arrays and multichannel electronics

    University essay from Lunds universitet/Avdelningen för Biomedicinsk teknik

    Author : Svante Rosenlind; [2023]
    Keywords : Focusing algorithms; k-wave; Spatiotemporal inverse filter; Gerchberg- Saxton algorithm; Focused ultrasound; Technology and Engineering;

    Abstract : High intensity focused ultrasound is a growing technique for tissue ablation, among other uses, and given its destructive capabilities, there is a need for control of where the energy is delivered. There exist a number of methods for focusing such ultrasound arrays, but these often assume prior knowledge of the impulse response, or require extensive full-system simulations. READ MORE