Learning latent space representations of high-dimensional world states has been at the core of recent rapid growth in reinforcement learning(RL). At the same time, RL algo- rithms have suffered from ignored uncertainties in the predicted estimates of model-free or model-based methods. In our work, we investigate both of these aspects...
Reinforcement learning has emerged as a popular tool for solving control tasks, with multiple works focusing on the complex and dynamic task of locomotion. However, the naive application of reinforcement learning to this problem often produces maladaptive policies that exploit the model or reward function. This results in behavior that...
This dissertation explores algorithms for learning ranking functions to efficiently solve search problems, with application to automated planning. Specifically, we consider the frameworks of beam search, greedy search, and randomized search, which all aim to maintain tractability at the cost of not guaranteeing completeness nor optimality. Our learning objective for...
Papers proposing novel machine learning algorithms tend to present the algorithm or technique in question in the best possible light. The standard practice is generally for authors to emphasize their proposed algorithms' performance in the precise setting where it is maximally impressive, often by only fully evaluating their best known...
Navigating the transition to adulthood is fraught with challenges, especially for those coming of age during an economic crisis. Individuals often rely on support from existing social networks, and establish social capital through deepening and expanding social ties, as they establish footholds in roles marking adulthood. Yet investing in social...
In light of broadening racial/ethnic educational attainment gaps, growing diversity within young populations, and the increasingly critical relationship between higher education and outcomes in adulthood, understanding the factors that shape educational attainment is an especially important task for research. With this goal in mind, the current study examined how national...
A continuous process for the production of ceramic materials has been studied.
This method reacted metal and oxidizer in a diffusionally controlled process,
demonstrated by reacting a magnesium particle stream and hot water vapor. Many small
rich hydrogen/air diffusion flames provide an atmosphere of hot water vapor, hydrogen,
and nitrogen...
Executive functions (including attentional shifting/flexibility, working memory, and inhibitory control) are strong predictors of children's early school success (Blair & Razza, 2007; Espy et al., 2004). The current study explored questions related to measurement of executive functions in preschool-aged children. Convergent and predictive validity were assessed for two traditional executive...
Front line staff members in youth correctional facilities are generally the primary adults in the lives of incarcerated young people. In many ways, they serve as surrogate parents who may promote positive development while holding delinquent youth accountable for their actions. They accomplish these often conflicting goals through the provision...
This thesis focuses on the economic efficiency of alternative habitat, streamfiow, and land use
allocations within a major drainage of the Pacific Northwest. Specifically, productivity of present habitat
and streamfiow conditions is evaluated with respect to anadromous fish species (chinook salmon and
steelhead trout) in the John Day River Basin,...