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 24 for “"Intrinsic Reward"”.

  1. Comparative Analysis of Intrinsic Reward-Based Reinforcement Learning Algorithms

    … often struggle in tasks with sparse or delayed rewards, since they receive little guidance about which actions to pursue. This thesis investigates how adding intrinsic rewards can help address that issue. We focus on three main methods: Count-Based bonuses, where states are hashed and infrequent …

    wustl Repository record for Comparative Analysis of Intrinsic Reward-Based Reinforcement Learning Algorithms (opens in a new tab)

  2. Reimagining Exploration: Theoretical Insights and Practical Advancements in Policy Gradient Methods

    … focus on policy parameterization choices and reward-shaping methods with intrinsic exploration bonuses. First, we analyze the influence of policy stochasticity on the optimization process. We formulate direct policy optimization within the optimization-by-continuation framework, which involves …

    liege Repository record for Reimagining Exploration: Theoretical Insights and Practical Advancements in Policy Gradient Methods (opens in a new tab) Download PDF for Reimagining Exploration: Theoretical Insights and Practical Advancements in Policy Gradient Methods (opens in a new tab)

  3. Full-time community college faculty with doctorates

    … included feelings of self-satisfaction and intrinsic reward in addition to "fit" with personal and family needs. Professional motivators included enjoyment of teaching along with the existing challenges of teaching in the community college.

    wm-mary Repository record for Full-time community college faculty with doctorates (opens in a new tab)

  4. Directed Exploration In Deep Reinforcement Learning Through Optimism

    … and based on learning from positive or negative rewards. One of the main dilemmas in any Reinforcement Learning task is the exploration-exploitation tradeoff: The question of when the agent should exploit known rewards and when it should take actions that it believes are suboptimal to learn more …

    queens Repository record for Directed Exploration In Deep Reinforcement Learning Through Optimism (opens in a new tab)

  5. Leader Self-Disclosure Depth: The Consequences of Leader’s Personal Disclosure for Followers

    … disclosers with many advantages such as higher intrinsic reward and improved interpersonal relations. Previous research on self-disclosure has mostly been developed within the context of close relationships (e.g., intimate partners) or strangers (e.g., who are not expected to interact again), …

    washington Repository record for Leader Self-Disclosure Depth: The Consequences of Leader’s Personal Disclosure for Followers (opens in a new tab)

  6. Reinforcement learning at the edge of chaos

    … updates pre-trained policies with a stabilising intrinsic reward function. This approach improves global stability without retraining from scratch and is compatible with any existing reinforcement learning controllers. Together, these contributions tackle a key limitation in current reinforcement …

    glasgow Repository record for Reinforcement learning at the edge of chaos (opens in a new tab) Download PDF for Reinforcement learning at the edge of chaos (opens in a new tab)

  7. ROBUST COUNTERFACTUAL LEARNING FOR CLINICAL DECISION-MAKING USING ELECTRONIC HEALTH RECORDS

    … and limited knowledge of the clinician's intrinsic reward function and environment dynamics. This motivates the need to tackle uncertainty and enable improved clinical policy generalization via context-based policy learning. We propose counterfactual frameworks to tackle the highlighted …

    gatech Repository record for ROBUST COUNTERFACTUAL LEARNING FOR CLINICAL DECISION-MAKING USING ELECTRONIC HEALTH RECORDS (opens in a new tab)

  8. Flow as a metastable brain state: evidence from EEG microstates, entropy-based complexity, and electromagnetic modulation

    … by deep absorption, effortless control, and intrinsic reward, yet it’s neurophysiological mechanisms remain incompletely understood. This thesis integrates psychological theory, electrophysiology, and electromagnetic field (EMF) models of brain function to conceptualize flow as a metastable …

    laurentian Repository record for Flow as a metastable brain state: evidence from EEG microstates, entropy-based complexity, and electromagnetic modulation (opens in a new tab)

  9. Knowledge workers and their relationships with organisations

    … but also that job challenge, problem solving, intrinsic reward and team identification were of importance to both groups of knowledge workers. This thesis presents an examination of knowledge workers’ perceptions of work, organisations and people management practices in the granulation and …

    aston Download PDF for Knowledge workers and their relationships with organisations (opens in a new tab)

  10. Risk homeostasis theory in simulated environments

    … is based on the simulation of physical risk with intrinsic reward rather than a 'points pay-off'. The second aim of the thesis is to examine a number of predictions made by RHT through the use of this alternative methodology. Since the pseudo-experimental design and post hoc analysis of …

    aston Download PDF for Risk homeostasis theory in simulated environments (opens in a new tab)

  11. Negotiating kendo capital and gendered identity in a Japanese sports university kendo club

    … a ‘bigendered identity’ and experience intrinsic reward in kendo more so than men as they are bound to a separate sphere of oppressive systems and articulations. Albeit a result of patriarchy, women do experience agency and can positively create their own raison d'être through the …

    vu-aus Download PDF for Negotiating kendo capital and gendered identity in a Japanese sports university kendo club (opens in a new tab)

  12. A comparison of relationships between older adults and their close friends and adult children.

    … including strength of the relationship, intrinsic and utilitarian rewards, and relationship strain, and to examine the differential impact of these variables on life satisfaction. Subjects were 44 widowed women over 65 years of age, who were retired and had at least one adult child not …

    nodak Repository record for A comparison of relationships between older adults and their close friends and adult children. (opens in a new tab)

  13. Modulating Reinforcement- Learning Parameters Using Agent Emotions

    … a computational model of emotion to provide an intrinsic reward function, the task can be shifted to optimisation of emotional feedback, allowing more high-level goals to be defined. While of most interest in a general, not necessarily competitive, social setting on a continuing task, such a …

    chalmers Repository record for Modulating Reinforcement- Learning Parameters Using Agent Emotions (opens in a new tab)

  14. Causality-inspired reinforcement learning: state abstractions, exploration, and representations

    … affect the transition of each factor) and in reward functions (which state factors determine task success). These causal structures enable RL agents to learn accurate dynamics and reward models, and to design hierarchical policies for solving complex, long-horizon tasks sample-efficiently …

    texas Repository record for Causality-inspired reinforcement learning: state abstractions, exploration, and representations (opens in a new tab)

  15. Motives and Values Associated with Participation in Intercollegiate Student-Athlete Community Service: Implications for Athletics Department Leadership

    … by being required, motivation to volunteer with intrinsic reward, motivation for career experience, and motivation through participation in a group/organization. These motivation factors were used as dependent variables and statistically significant relationships occurred when comparing …

    arizona-thes Repository record for Motives and Values Associated with Participation in Intercollegiate Student-Athlete Community Service: Implications for Athletics Department Leadership (opens in a new tab)

  16. Reinforcement Learning Based Systems Modeling and Decision-Making

    … enhance the model performance, the concept of intrinsic motivation is introduced into the proposed framework by proposing a unique intrinsic reward. Second, we focus on the droplet pinch-off behaviors which affect the quality of final printed products in IJP in Chapter 3. Specifically, we …

    buffalo Repository record for Reinforcement Learning Based Systems Modeling and Decision-Making (opens in a new tab)

  17. Learning navigation policies with deep reinforcement learning

    … in learning control policies in terminal reward settings, as this scenario requires the least amount of human priors and would thus largely automate the training of artificial decision-making agents. As structured and guided exploration becomes vital in this case, we again question the …

    freiburg-diss Repository record for Learning navigation policies with deep reinforcement learning (opens in a new tab)

  18. Towards Optimistic, Imaginative, and Harmonious Reinforcement Learning in Single-Agent and Multi-Agent Environments

    … agent’s underestimation of the expected future reward, i.e. the state-value function. To alleviate the issue we propose to enhance the agent’s state-value function approximator with more global information. In visual navigation, we do so by learning the spatio-temporal relationship between …

    adelaide Repository record for Towards Optimistic, Imaginative, and Harmonious Reinforcement Learning in Single-Agent and Multi-Agent Environments (opens in a new tab)

  19. Obedience-based Multi-Agent Cooperation for Sequential Social Dilemmas

    … Reinforcement Learning (MARL) settings by intrinsically rewarding agents for obeying the commands of other agents. At every timestep, agents exchange commands through a cheap-talk channel. During the following timestep, agents are rewarded both for taking actions that conform to commands …

    waterloo Repository record for Obedience-based Multi-Agent Cooperation for Sequential Social Dilemmas (opens in a new tab)

  20. Exploration of dentists’ characteristics associated with caring for disadvantaged patients using a mixed-methods approach.

    … relieving pain and improving their wellness; 4) ‘Intrinsic reward’; the personal gain in receiving simple, unexpected rewards that made a difference; 5) ‘Resilience’; the overarching theme, derived from personal experiences and challenges of the work environment. In Stage 2, 1523 questionnaires …

    adelaide Repository record for Exploration of dentists’ characteristics associated with caring for disadvantaged patients using a mixed-methods approach. (opens in a new tab)

Page 1 of 2