Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 19 of 19 for “"F-DIVERGENCE"”.

  1. Policy Optimization with f-Divergence Regularization

    … algorithms based on the Kullback–Leibler (KL) divergence. In this thesis, we extend previous work by using a more general form of divergence, the f-divergence, and design a new family of algorithms that can improve the learning policy with theoretical improvement guarantees. Our method, …

    carleton Repository record for Policy Optimization with f-Divergence Regularization (opens in a new tab)

  2. Error and f -Divergence Analysis of Generative Adversarial Network

    … of error and estimation bounds of various f-divergence metrics in the context of generative adversarial networks (GANs). The error convergence rate is developed and explained by a class of functions with the discriminator and generator neural networks. The new class of functions is uniformly …

    mississippi Repository record for Error and f -Divergence Analysis of Generative Adversarial Network (opens in a new tab)

  3. Empirical risk minimization with f-divergence regularization for machine learning

    … Building on this insight, a broad family of f-divergences is introduced as potential regularizers for empirical risk minimization. Under mild assumptions, solutions for general f-divergences are derived, and the concept of the normalization function is formally defined. Furthermore, a dual …

    whiterose Download PDF for Empirical risk minimization with f-divergence regularization for machine learning (opens in a new tab)

  4. ΣΥΜΒΟΛΗ ΣΤΗΝ ΑΝΑΛΥΣΗ ΚΑΤΗΓΟΡΙΚΩΝ ΔΕΔΟΜΕΝΩΝ

    … MODELS AND ASYMMETRY MODELS WITH THE HELP OF F-DIVERGENCE AND THE CONSTRUCTION OF THE FAMILIES OF F-DIVERGENCE ASSOCIATION MODELS AND F-DIVERGENCE ASYMMETRY MODELS RESPECTIVELY, WHICH INCLUDE THE CLASSICAL ASSOCIATION AND ASYMMETRY MODELS AS SPECIAL CASES.- (CH. 4& 5). (IV) THE GENERALIZATION OF …

    greece Repository record for ΣΥΜΒΟΛΗ ΣΤΗΝ ΑΝΑΛΥΣΗ ΚΑΤΗΓΟΡΙΚΩΝ ΔΕΔΟΜΕΝΩΝ (opens in a new tab)

  5. Generative Adversarial Networks Based on a General Parameterized Family of Generator Loss Functions

    … in terms of the so-called Jensen-$f$-divergence, a natural generalization of the Jensen-Shannon divergence to the $f$-divergence. We also show that our result recovers as special cases several GANs from the literature, including the original GAN, least square GAN (LSGAN), $\alpha$-GAN …

    queens Repository record for Generative Adversarial Networks Based on a General Parameterized Family of Generator Loss Functions (opens in a new tab)

  6. The formal definition of reference priors under a general class of divergence

    … in the contexts under Kullback-Leibler divergence. In special cases with common support and other regularity conditions, Ghosh, Mergel and Liu (2011) derived a general f-divergence criterion for prior selection. We generalize Ghosh, Mergel and Liu's (2011) results to the case without …

    missouri Repository record for The formal definition of reference priors under a general class of divergence (opens in a new tab)

  7. Derivative pricing and logarithmic portfolio optimization in incomplete markets

    … special distances but the whole class of f-divergence distances defined by strictly convex, differentiable functions. <br>Another problem studied in this thesis is the determination of optimal portfolios for logarithmic utility in general semimartingale models. The solution is given …

    freiburg-diss Repository record for Derivative pricing and logarithmic portfolio optimization in incomplete markets (opens in a new tab)

  8. Visual Learning Using Synthetic Data

    … adversarial framework that minimizes a chosen f-divergence and is both practical and theoretically motivated. We also show how to deal with complications arising in this framework, such as complex optimization dynamics and label-distribution mismatch. We demonstrate the effectiveness of our …

    toronto-retro Repository record for Visual Learning Using Synthetic Data (opens in a new tab)

  9. Variational Information-Theoretic Atoms-in-Molecules

    … framework. This characterizes the family of f-divergence measures as necessary and sufficient for deriving Hirshfeld scheme. Then, we developed a variational version of Hirshfeld partitioning method, called Additive Variational Hirshfeld (AVH). The key idea is finding the promolecular density, …

    mcmaster Repository record for Variational Information-Theoretic Atoms-in-Molecules (opens in a new tab)

  10. GEOMETRIC APPROACHES TO DISCRIMINATIVE TRAINING

    … of NGD. The loss functions, defined by Bregman divergence, generalizes the log-likelihood objective and can be easily coupled with NGD for optimization. The proposed framework is flexible, allowing us to choose proper convex functions that leads to better training performance. Finally, …

    jhu Repository record for GEOMETRIC APPROACHES TO DISCRIMINATIVE TRAINING (opens in a new tab)

  11. Odhady diskrétního rozložení pravděpodobnosti a bootstrap

    Dizertační práce se zabývá netradičními odhady diskrétního rozdělení pravděpodobnosti kategoriální veličiny z jejích pozorovaných hodnot. Při odhadech se využívá gradientu kvazinormy tohoto rozdělení pravděpodobnosti a tzv. přímkového odhadu. Zpřesnění gradientních odhadů bylo provedeno metodou …

    brno-tech Repository record for Odhady diskrétního rozložení pravděpodobnosti a bootstrap (opens in a new tab)

  12. DATA-DRIVEN OPTIMIZATION UNDER UNCERTAINTY IN THE ERA OF BIG DATA AND DEEP LEARNING: GENERAL FRAMEWORKS, ALGORITHMS, AND APPLICATIONS

    … deep generative adversarial network (GAN), an f-divergence-based ambiguity set of wind power distributions is constructed as a ball in the probability space centered at the distribution induced by a generator neural network. To facilitate its solution process, the resulting distributionally …

    cornell Repository record for DATA-DRIVEN OPTIMIZATION UNDER UNCERTAINTY IN THE ERA OF BIG DATA AND DEEP LEARNING: GENERAL FRAMEWORKS, ALGORITHMS, AND APPLICATIONS (opens in a new tab)

  13. Odhady diskrétních rozdělení pravděpodobnosti pro aplikace

    Diplomová práce je zaměřena na řešení statistického problému nalezení rozdělení pravděpodobnosti diskrétní náhodné veličiny na základě pozorovaných dat. Tyto odhady jsou získány minimalizací pseudokvazinormy, která je zde zavedena. Práce se zabývá vlastnostmi této pseudokvazinormy. Obsahuje také …

    brno-tech Repository record for Odhady diskrétních rozdělení pravděpodobnosti pro aplikace (opens in a new tab)

  14. Kvazinormy diskrétních rozdělení pravděpodobnosti a jejich aplikace

    Disertační práce je zaměřena na řešení statistického problému nalezení rozdělení pravděpodobnosti diskrétní náhodné veličiny na základě pozorovaných dat. Tyto odhady jsou získány minimalizací kvazinorem za předem daných omezujících podmínek kladených na hledaná rozdělení. Práce se dále zabývá …

    brno-tech Repository record for Kvazinormy diskrétních rozdělení pravděpodobnosti a jejich aplikace (opens in a new tab)

  15. Variational approximation for importance sampling and statistical inference on social influence

    Monte Carlo methods are widely used in statistical computing area to solve different problems. Social network analysis plays an importance role in many fields. In this dissertation, we focus on improving the efficiency of importance sampling, detecting the degrees of influence in networks, and …

    uiuc Repository record for Variational approximation for importance sampling and statistical inference on social influence (opens in a new tab)

  16. The fundamental limits of statistical data privacy

    The Internet is shaping our daily lives. On the one hand, social networks like Facebook and Twitter allow people to share their precious moments and opinions with virtually anyone around the world. On the other, services like Google, Netflix, and Amazon allow people to look up information, watch …

    uiuc Repository record for The fundamental limits of statistical data privacy (opens in a new tab)

  17. Metoda bootstrap a její aplikace

    Diplomová práce popisuje metodu bootstrap a její použití pro určení přesnosti odhadu, tvorbu konfidenčních intervalů a testování statistických hypotéz. Dále je předložena metoda odhadu diskrétního rozdělení pravděpodobnosti kategoriální veličiny využívající gradient kvazinormy tohoto rozdělení. …

    brno-tech Repository record for Metoda bootstrap a její aplikace (opens in a new tab)

  18. Fitování rozdělení pravděpodobnosti pro aplikace

    Diplomová práce popisuje metodu bootstrap a její použití pro tvorbu konfidenčních intervalů, při testování statistických hypotéz a v regresní analýze. Představujeme konfidenční interval pro individuální hodnotu. Dále se zaobíráme metodou odhadu diskrétního rozdělení pravděpodobnosti kategoriální …

    brno-tech Repository record for Fitování rozdělení pravděpodobnosti pro aplikace (opens in a new tab)