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 20 of 45 for “"Regret Minimization"”.

  1. Regret Minimization And Related Decision Rules

    … approach to decision making-minimizing expected regret-to obtain minimax weighted expected regret (MWER). We provide an axiomatization that characterizes preferences induced by MWER both in the static and dynamic case. This same concept of weighted probability distributions can also be applied to …

    cornell Repository record for Regret Minimization And Related Decision Rules (opens in a new tab)

  2. Digital marketplace platform inventory management under disruption : a robust optimization/ regret minimization approach

    … an alternative framework, namely the maximum regret minimization (MRM), which is less conservative than the RO. The column-and-constraint generation method is used to test the proposed models under various scenarios with different demand correlation levels. As a method of solving these …

    ubc Repository record for Digital marketplace platform inventory management under disruption : a robust optimization/ regret minimization approach (opens in a new tab) Download PDF for Digital marketplace platform inventory management under disruption : a robust optimization/ regret minimization approach (opens in a new tab)

  3. Minimax Regret Optimization with Applications in Healthcare Operations

    … This thesis explores the application of the regret minimization model to deal with demand uncertainty in a home health care location problem and uncertainty of surgery duration in an operating room planning problem. The objective is to find a balance between being conservative and being …

    ubc Repository record for Minimax Regret Optimization with Applications in Healthcare Operations (opens in a new tab) Download PDF for Minimax Regret Optimization with Applications in Healthcare Operations (opens in a new tab)

  4. Metody pro hraní hry Liar's Dice s využítím dynamického programování

    … nad algoritmy Q-Learning a Counterfactual Regret Minimization. Ve výsledku bylo dosaženo úspěšnosti 69,147 % ve hře proti Q-Learning a úspěšnosti pouze 25 % proti algoritmu Counterfactual Regret Minimization. Tato práce poskytuje hlavní přehled o tom, jak upravená verze algoritmu SARSA je …

    brno-tech Repository record for Metody pro hraní hry Liar's Dice s využítím dynamického programování (opens in a new tab)

  5. Metody pro hraní hry 'Liar's Dice' s využítím dynamického programování

    … s algoritmy Q-Learning a Counterfactual Regret Minimization. Podařilo se mi dosáhnou úspěšnosti 69,147\,\% ve hře proti Q-Learning a úspěšnosti pouze 29,84\,\% proti algoritmu Counterfactual Regret Minimization. Hlavním zjištěním této práce je, že upravená verze SARSA algoritmu je velmi …

    brno-tech Repository record for Metody pro hraní hry 'Liar's Dice' s využítím dynamického programování (opens in a new tab)

  6. Accommodating Exogenous Variable and Decision Rule Heterogeneity in Discrete Choice Models: Application to Bicyclist Route Choice

    … the random utility framework along with random regret minimization approach. Further, instead of assuming the number of segments (as 2), we conduct an exhaustive exploration with multiple segments across the two decision rules. Within each segment we also allow for unobserved heterogeneity. The …

    ucf Repository record for Accommodating Exogenous Variable and Decision Rule Heterogeneity in Discrete Choice Models: Application to Bicyclist Route Choice (opens in a new tab)

  7. Strojové hraní karetní hry "Královské rošády" metodami umělé inteligence

    … Monte Carlo Tree Search (MCTS) a Counterfactual Regret Minimization (CFR) a následně zhodnotit jejich výkonnost proti lidským hráčům. Implementované algoritmy byly řádně testovány proti náhodnému hráči a lidským oponentům. Výsledky ukazují, že MCTS je pro řešení tohoto problému efektivnější než …

    brno-tech Repository record for Strojové hraní karetní hry "Královské rošády" metodami umělé inteligence (opens in a new tab)

  8. Improved worst-case regret bounds for randomized least-squares value iteration

    This work studies regret minimization with randomized value functions in reinforcement learning. In tabular finite-horizon Markov Decision Processes, we introduce a clipping variant of one classical Thompson Sampling (TS)-like algorithm, randomized least-squares value iteration (RLSVI). Our …

    uiuc Repository record for Improved worst-case regret bounds for randomized least-squares value iteration (opens in a new tab)

  9. Temporal aspects of adaptive online learning: continuity and representation

    … Convex Optimization, achieving optimal static regret bounds in the vanilla setting and its variant with switching costs. The benefits are extended to another classical online learning problem called Learning with Expert Advice. The second part of the dissertation focuses on temporal …

    bu Repository record for Temporal aspects of adaptive online learning: continuity and representation (opens in a new tab)

  10. Hraní karetní hry Poker počítačem s využitím metod strojového učení

    … řešení her se zaměřením na metodu Counterfactual Regret Minimization (CFR) a její varianty. Jsou zkoumány techniky abstrakce pro snížení složitosti hry. Implementace agenta využívá Monte Carlo CFR External Sampling s distribution-aware abstrakcí a je popsána a vyhodnocena prostřednictvím zápasů s …

    brno-tech Repository record for Hraní karetní hry Poker počítačem s využitím metod strojového učení (opens in a new tab)

  11. New Models qnd Algorithms for Bandits and Markets

    … a generic structural assumption on rewards to regret rates in an online optimization problem, is not fully developed. The primary goal of this dissertation, therefore, will be to fill out the space of models, algorithms, and assumptions used in sequential decision making problems. Toward this …

    penn Repository record for New Models qnd Algorithms for Bandits and Markets (opens in a new tab)

  12. Texas Hold’em poker by LLM

    … computational methods like Counterfactual Regret Minimization (CFR) Zinkevich et al. (2008). This research demonstrates how LLAMA2 can interactively improve its strategies through trial-and-error within a reinforcement learning environment, ultimately surpassing baseline methods in terms of …

    texas Repository record for Texas Hold’em poker by LLM (opens in a new tab)

  13. PERFORMANCE GUARANTEES FOR ONLINE LEARNING: CASCADING BANDITS AND ADVERSARIAL CORRUPTIONS

    … bandit model, wherein we design and analyze regret minimization (RM) and best arm identification (BAI) algorithms. We illustrate that our algorithms are nearly optimal by comparing their performance metrics with minimax lower bounds on the regrets and failure probabilities for any algorithm. …

    nus Repository record for PERFORMANCE GUARANTEES FOR ONLINE LEARNING: CASCADING BANDITS AND ADVERSARIAL CORRUPTIONS (opens in a new tab) Download file for PERFORMANCE GUARANTEES FOR ONLINE LEARNING: CASCADING BANDITS AND ADVERSARIAL CORRUPTIONS (opens in a new tab)

  14. Information structure in bandit learning

    … Bayesian best-arm identification, and regret minimization in Bayesian bandits. We also study a variant of Thompson sampling, a widely used posterior sampling algorithm, and establish new regret guarantees that capture the impact of prior information.

    texas Repository record for Information structure in bandit learning (opens in a new tab)

  15. A regularization perspective on spectral sparsification

    … connectionbetween spectral sparsification and regret minimization in online matrix games, i.e.,online convex programming over the positive semidefinite cone. While this connec-tion was previously noted [24, 47], we formally reduce graph sparsification to a matrixregret minimization problem, …

    bu Repository record for A regularization perspective on spectral sparsification (opens in a new tab)

  16. Online algorithm design via smoothing with application to online experiment selection

    … a systematic approach with a clear connection to regret minimization for design and worst case analysis of online optimization algorithms. We apply this approach to online experiment design problems. Our results on the convergence rate analysis of proximal gradient homotopy algorithm extends the …

    washington Repository record for Online algorithm design via smoothing with application to online experiment selection (opens in a new tab)

  17. Swap-regret-minimizing bandits for distributed network optimization

    … and analyze algorithms that not only minimize regret but also steer collective behavior toward equilibrium. The contributions of this dissertation are threefold. First, we propose a new framework based on swap-regret minimization and online mirror descent, and establish high-probability regret …

    uvic Repository record for Swap-regret-minimizing bandits for distributed network optimization (opens in a new tab)

  18. Predictable Sequences and Competing with Strategies

    … study online learning with an extended notion of regret, which is defined with respect to a set of strategies. We develop tools for analyzing the minimax rates and deriving efficient learning algorithms in this scenario. While the standard methods for minimizing the usual notion of regret fail, …

    penn Repository record for Predictable Sequences and Competing with Strategies (opens in a new tab)

  19. The Influence of Rich Media (Video) Versus Infographics on Consumer Valuation of Cheese Attributes: Evidence from a Non-Hypothetical Choice Experiment

    … Our findings hold under WTP-space estimation, regret minimization models, and balance tests. These findings show that media richness differentially amplifies attribute valuations, suggesting dairies use attribute tailored marketing strategies to boost revenue.</p>

    ut-knoxville Repository record for The Influence of Rich Media (Video) Versus Infographics on Consumer Valuation of Cheese Attributes: Evidence from a Non-Hypothetical Choice Experiment (opens in a new tab)

  20. Trustworthy interactive learning, a story of fairness, robustness and safety

    … (RKHS) and introduces novel algorithms for regret minimization and pure exploration tasks. This exploration extends to level set estimation in kernel bandits, shedding light on nearly optimal algorithms and their performance characteristics. Building on these foundational principles, the …

    washington Repository record for Trustworthy interactive learning, a story of fairness, robustness and safety (opens in a new tab)

Page 1 of 3