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 9 of 9 for “"Direct Policy Search"”.

  1. DIAGNOSTIC ASSESSMENT AND ADVANCEMENT OF MULTI-OBJECTIVE RESERVOIR CONTROL UNDER UNCERTAINTY

    … optimization techniques coupled with direct policy search can reliably and flexibly find suitable control policies that adapt to multi-sectorial water needs and to hydro-climatic uncertainty. The second study demonstrates the benefits of cooperative parallel MOEA architectures to …

    cornell Repository record for DIAGNOSTIC ASSESSMENT AND ADVANCEMENT OF MULTI-OBJECTIVE RESERVOIR CONTROL UNDER UNCERTAINTY (opens in a new tab)

  2. Shaping and policy search for nearest-neighbour control policies with applications to vehicle steering

    … algorithmically. There is a need for further research into motion control techniques to address this problem adequately for computer animation and robotics applications. In this thesis, we describe a novel method for the synthesis of compact control policies for a variety of motion control …

    ubc Repository record for Shaping and policy search for nearest-neighbour control policies with applications to vehicle steering (opens in a new tab) Download PDF for Shaping and policy search for nearest-neighbour control policies with applications to vehicle steering (opens in a new tab)

  3. Applications of Interpretable Data-driven Approaches in Climate, Hydrology, and Water Resource Management

    … offered new opportunities in hydro-climate research with tremendous data and highly complex, nonlinear models. However, a model tends to lose interpretability when it becomes complex, which can degrade its transferability under out-of-sample conditions. Therefore, the tradeoff between model …

    cornell Repository record for Applications of Interpretable Data-driven Approaches in Climate, Hydrology, and Water Resource Management (opens in a new tab)

  4. Can Modern Multi-Objective Evolutionary Algorithms Discover High-Dimensional Financial Risk Portfolio Tradeoffs for Snow-Dominated Water-Energy Systems?

    … in yearly revenue. Evolutionary multi-objective direct policy search (EMODPS) can be used to identify time adaptive stochastic rules that inform optimal financial decisions based on state and exogenous information. However, the resulting financial risk mitigation portfolio planning problem is …

    cornell Repository record for Can Modern Multi-Objective Evolutionary Algorithms Discover High-Dimensional Financial Risk Portfolio Tradeoffs for Snow-Dominated Water-Energy Systems? (opens in a new tab)

  5. Reinforcement learning in continuous state- and action-space

    … artificial neural network to approximate the policy function [21, 55]. The major contributions of this thesis include the investigation of the applicability of action selection by numerical optimization methods, including gradient-ascent along with other derivative-based and derivative-free …

    westminster Repository record for Reinforcement learning in continuous state- and action-space (opens in a new tab) Download PDF for Reinforcement learning in continuous state- and action-space (opens in a new tab)

  6. Decision making with inference and learning methods

    … for probabilistic inference. This allows us to directly apply a wide variety of popular, practical algorithms to these tasks. In Chapter 1 we provide an overview of the general problem of sequential decision making and a broad description of various solution methods. Much of the remaining work …

    ubc Repository record for Decision making with inference and learning methods (opens in a new tab) Download PDF for Decision making with inference and learning methods (opens in a new tab)

  7. A Framework for Aggregation of Multiple Reinforcement Learning Algorithms

    … dynamic learning problems without the need to search for the optimal learning algorithm or the optimal values of learning parameters. It is claimed that several complementary learning algorithms can be integrated in AMRLS to improve the learning performance in terms of success rate, robustness, …

    waterloo Repository record for A Framework for Aggregation of Multiple Reinforcement Learning Algorithms (opens in a new tab)

  8. Policy Search Based Relational Reinforcement Learning using the Cross-Entropy Method

    … RRL algorithm named Cerrla that creates policies directly from a set of learned relational “condition-action” rules using the Cross-Entropy Method (CEM) to control policy creation. The CEM assigns each rule a sampling probability and gradually modifies these probabilities such that the randomly …

    waikato-masters Repository record for Policy Search Based Relational Reinforcement Learning using the Cross-Entropy Method (opens in a new tab) Download file for Policy Search Based Relational Reinforcement Learning using the Cross-Entropy Method (opens in a new tab)

  9. Deep Reinforcement Learning for the Control of Energy Storage in Grid-Scale and Microgrid Applications

    The European and worldwide directives and targets for renewable energy integration, motivated by the imminent need to decarbonize the electricity sector, are imposing severe changes to the conventional electrical power system. The inherent unpredictability of the instantaneous energy production …

    liege Repository record for Deep Reinforcement Learning for the Control of Energy Storage in Grid-Scale and Microgrid Applications (opens in a new tab) Download PDF for Deep Reinforcement Learning for the Control of Energy Storage in Grid-Scale and Microgrid Applications (opens in a new tab)