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 98 for “"State Representation"”.

  1. Dynamic State Representation for Homeostatic Agents

    … For complex environments, the explorable state space can easily become unmanageable, and efficient approximations are needed. The Generic Animat model (GA model), heavily influenced by biology, takes an approach utilising a dynamic graph to represent the state. This thesis is part of the …

    chalmers Repository record for Dynamic State Representation for Homeostatic Agents (opens in a new tab)

  2. Learning Action-State Representation Forests for Implicitly Relational Worlds

    … execution from scalar rewards based on agent state and action. Relational reinforcement learning (RRL) combines these two fields. In this dissertation, I present an RRL technique emphasizing relations that are merely implicit in multidimensional, continuous object attributes, such as position, …

    oklahoma-thes Repository record for Learning Action-State Representation Forests for Implicitly Relational Worlds (opens in a new tab) Restricted access

  3. GRLSR : a goal-conditioned RL related state representation learning method

    In the field of reinforcement learning (RL), representation learning has been proven to be a useful tool in robotic tasks. However, many representation learning methods for RL tasks are developed without explicit objectives linked to RL theory. The efficiency of representation learning for …

    texas Repository record for GRLSR : a goal-conditioned RL related state representation learning method (opens in a new tab)

  4. Learning from Pixels: Image-Centric State Representation Reinforcement Learning for Goal Conditioned Surgical Task Automation

    … share a common reliance on accurate ground truth state observations. This reliance poses a substantial hurdle, particularly in real-world scenarios where capturing an accurate state representation becomes notably challenging.This study addresses the aforementioned challenge by exploiting an …

    umn Repository record for Learning from Pixels: Image-Centric State Representation Reinforcement Learning for Goal Conditioned Surgical Task Automation (opens in a new tab)

  5. Establishing object-state representation in language comprehension: Evidence from picture verification, eye-tracking and ERPs

    … this thesis explored when and how object-state representation is established, maintained, and retrieved in language comprehension. We firstly examined whether different object-state representations could be established under two contrasting linguistic contexts (e.g., no change – “choose …

    whiterose Download PDF for Establishing object-state representation in language comprehension: Evidence from picture verification, eye-tracking and ERPs (opens in a new tab)

  6. Examining Space Usage for Developing a Current and Future State Representation of Patrons Needs: A Case Study in H.M. Briggs Library

    … Library on the main campus of South Dakota State University. The goal of this project is to gather data that can be used for redesigning spaces in H. M. Briggs Library to better meet the needs of patrons. To redesign the space, it is necessary to understand current usage of library space and …

    sdstate Repository record for Examining Space Usage for Developing a Current and Future State Representation of Patrons Needs: A Case Study in H.M. Briggs Library (opens in a new tab)

  7. Graph-ReCoN: A Graph-Based Framework for Induced Relational Context in Model-Free Reinforcement Learning

    … distinct, they share a common dependency – the state representation, i.e., how the agent views and encodes its environment. State Representation Learning (SRL), a fundamental component of Model-Free RL, offers a variety of methods that seek to do just that – enhance the state representation. …

    gatech Repository record for Graph-ReCoN: A Graph-Based Framework for Induced Relational Context in Model-Free Reinforcement Learning (opens in a new tab)

  8. Volterra Systems with Realizable Kernels

    We compare an internal state method and a direct Runge-Kutta method for solving Volterra integro-differential equations and Volterra delay differential equations. The internal state method requires the kernel of the Volterra integral to be realizable as an impulse response function. We discover …

    vt Repository record for Volterra Systems with Realizable Kernels (opens in a new tab)

  9. Understanding representation learning for deep reinforcement learning

    <p>Representation learning is essential to practical success of reinforcement learning. Through a state representation, an agent can describe its environment to efficiently explore the state space, generalize to new states and perform credit assignment from delayed feedback. These representations …

    oxford Repository record for Understanding representation learning for deep reinforcement learning (opens in a new tab)

  10. Toward General Techniques for Autonomous Rendezvous

    … transfer, investigates the use of a new orbit state representation for this purpose, and finally introduces and tests a method for executing orbit transfer and rendezvous using this orbit representation. Existing methods have used primer vector theory as a basis for indirect optimization and …

    cornell Repository record for Toward General Techniques for Autonomous Rendezvous (opens in a new tab)

  11. From Demonstration to Dynamic Interaction: Enabling Long-Term Robotic Planning

    … and the difficulty of obtaining accurate state representations. This thesis explores how we can advance these methods to enable robust performance in real-world, unstructured environments. We begin by exploring how to redefine state representation, presenting two complementary approaches. …

    maryland Repository record for From Demonstration to Dynamic Interaction: Enabling Long-Term Robotic Planning (opens in a new tab)

  12. Theoretical studies on liquid water

    Water in its liquid state is known for possessing properties that run contrary to those of most liquids. These properties arise from a change in distance of the non-hydrogen bonded neighbor. The changes in distance are a result of changes in local structure from a tetrahedral structure (type-I …

    ttu Repository record for Theoretical studies on liquid water (opens in a new tab)

  13. Abstractions and transformations for automated data network configuration

    … to an intermediate network-wide configuration state representation. A second transformation step converts the intermediate representation into the low-level device configuration state appropriate for the target device, which is then assembled using simple templates. We show how this multi-stage …

    adelaide Repository record for Abstractions and transformations for automated data network configuration (opens in a new tab)

  14. A unified framework for disentangled 3D mouse body-state estimation

    … of video-derived measurements depends on the representation used to describe the animal. Modern markerless methods can recover 2D/3D landmarks, silhouettes, 3D surfaces, or learned behavioural embeddings. These representations are useful but incomplete descriptions of body state: sparse …

    ubc Repository record for A unified framework for disentangled 3D mouse body-state estimation (opens in a new tab) Download PDF for A unified framework for disentangled 3D mouse body-state estimation (opens in a new tab)

  15. Path integral formulation of dissipative quantum dynamics

    … real-time path integral technique in coherent-state representation. This method is applied to a damped harmonic oscillator within the Caldeira-Leggett model. In order to get the stationary trajectories the corresponding Lagrangian function is diagonalized and then the path integrals are …

    qucosa-diss

  16. Contributions to deep reinforcement learning and its applications in smartgrids

    … increasing the asymptotic bias, a smaller state representation decreases the risk of overfitting. An original theoretical contribution relies on expressing the quality of a state representation by bounding $L_1$ error terms of the associated belief states. We also discuss and empirically …

    liege Repository record for Contributions to deep reinforcement learning and its applications in smartgrids (opens in a new tab) Download PDF for Contributions to deep reinforcement learning and its applications in smartgrids (opens in a new tab)

  17. Assembling Information from Big Corpora by Focusing Machine Reading

    … (FR), as a Markov Decision Process, whose state-representation is comprised of domain-agnostic features related to the current state of the search and the dynamics of the search process. We employ reinforcement learning (RL) to find a policy to search for multi-hop connections in the …

    arizona-thes Repository record for Assembling Information from Big Corpora by Focusing Machine Reading (opens in a new tab)

  18. Haag Duality in the Thermodynamic Limit of the Toric Code

    … Hamiltonians for the toric code and their ground states, showing that Haag duality holds for cones in the corresponding ground state representation. That is, given a cone Λ, the von Neumann algebras R_Λ and R_Λ_c of observables localized in Λ and Λ_c respectively are each other’s commutant. We …

    carleton Repository record for Haag Duality in the Thermodynamic Limit of the Toric Code (opens in a new tab)

  19. Applications of a Quasi-Diabatic Hamiltonian Method to Problems in Nonadiabatic Chemistry

    A method for computing coupled, diabatic state representations of the lowest electronic states coupled by conical intersections of polyatomic systems featuring large molecular motion, such as dissociation, is applied to the photodissociation of CH2OH from the first and second excited states. …

    jhu Repository record for Applications of a Quasi-Diabatic Hamiltonian Method to Problems in Nonadiabatic Chemistry (opens in a new tab)

  20. Autonomous Drone for Dynamic Smoke Plume Tracking

    … approach uses binary smoke segmentation masks as state representation, enabling robust adaptation without explicit flow measurements. Evaluation in a high-fidelity Unreal Engine simulation and real-world field experiments demonstrates that the DRL-based controller exhibits superior robustness and …

    umn Repository record for Autonomous Drone for Dynamic Smoke Plume Tracking (opens in a new tab)

Page 1 of 5