Essays about: "entropy regularization"

Showing result 1 - 5 of 7 essays containing the words entropy regularization.

  1. 1. Mitigating Unintended Bias in Toxic Comment Detection using Entropy-based Attention Regularization

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

    Author : Fabio Camerota; [2023]
    Keywords : XLNet; BERT; Toxic Comment Classification; Entropy-based Attention Regularization; XLNet; BERT; Toxisk Kommentar Klassificering; Entropibaserad uppmärksamhetsreglering;

    Abstract : The proliferation of hate speech is a growing challenge for social media platforms, as toxic online comments can have dangerous consequences also in real life. There is a need for tools that can automatically and reliably detect hateful comments, and deep learning models have proven effective in solving this issue. READ MORE

  2. 2. Improving the Robustness of Deep Neural Networks against Adversarial Examples via Adversarial Training with Maximal Coding Rate Reduction

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

    Author : Hsiang-Yu Chu; [2022]
    Keywords : Machine learning; Deep neural networks; Loss function; Adversarial example; Adversarial attack; Adversarial training; Maskininlärning; Djupa neurala nätverk; Förlustfunktion; Motståndarexempel; Motståndarattack; Motståndsträning;

    Abstract : Deep learning is one of the hottest scientific topics at the moment. Deep convolutional networks can solve various complex tasks in the field of image processing. However, adversarial attacks have been shown to have the ability of fooling deep learning models. READ MORE

  3. 3. Regularized Fine-tuning Strategies for Neural Language Models : Application of entropy regularization on GPT-2

    University essay from Uppsala universitet/Institutionen för lingvistik och filologi

    Author : Jae Eun Hong; [2022]
    Keywords : text generation; entropy regularization; neural language model; gpt-2;

    Abstract : Deep neural language models like GPT-2 is undoubtedly strong at text generation, but often requires special decoding strategies to prevent producing degenerate output - namely repetition. The use of maximum likelihood training objective results in a peaked probability distribution, leading to the over-confidence of neural networks. READ MORE

  4. 4. Aortic Segmentation in Real-Time Flow Exercise Cardiac Magnetic Resonance Images using Convolutional Neural Networks

    University essay from Lunds universitet/Matematik LTH

    Author : Mathilda Larsson; [2021]
    Keywords : Mathematics and Statistics;

    Abstract : Background: Cardiovascular disease is common and it is therefore important to have tools to investigate pathophysiology and normal physiology of the cardiovascular system. One such tool is using cardiovascular magnetic resonance in order to measure blood flow in vessels, particularly while the participant imaged is exercising. READ MORE

  5. 5. Reconstruction in Computerized Tomography using Entropy Regularization

    University essay from KTH/Skolan för teknikvetenskap (SCI)

    Author : Theoharis Charitidis; Martin Sjöberg; [2017]
    Keywords : ;

    Abstract : Inverse problems are a class of problems that often appear in science. They canbe thought of as calculating the cause that corresponds to measured data, andit is the opposite (i.e. inverse) of a direct problem, where data is calculated froma known source. READ MORE