Washington University Medical Center
Washington University St. Louis: Open ScholarshipNot a member yet
26344 research outputs found
Sort by
A Control Theoretic Approach to the Stochastic Multi-armed Bandit Problem With Applications in Hyperparameter Optimization
No full textDecision-making under uncertainty is a fundamental problem encountered frequently in many real-world applications. This challenge has been rigorously formulated as the Stochastic Multi-Armed Bandit (SMAB) problem, which consists of a learner interacting with an environment. For each interaction, the learner selects an action and then receives a reward from the environment based on the chosen action. The learner\u27s objective is to maximize the accumulated reward over a set number of rounds. This thesis addresses the SMAB problem by leveraging the field of Control Theory and dynamical systems. We specifically focus on a SMAB environment where the rewards are the output of a Linear Gaussian Dynamical System (LGDS). The core contribution of this thesis is to demonstrate how Control Theory can enhance our current understanding of decision-making under uncertainty through the SMAB problem. We address two different directions relevant to the SMAB problem using a control theoretic approach. The first direction is the issue of how to efficiently explore the environment\u27s action space. We discover that a LGDS property called observability, which measures the difficulty of estimating the LGDS\u27s state variable, can be utilized to increase the amount of information gained during exploration. The second direction we consider is how to utilize the environmental structure to predict each action\u27s reward more effectively. We show that a representation of the Kalman filter, where the Kalman filter is the optimal one-step predictor of the LGDS\u27s output in the mean-squared error sense, can be extracted for predicting each action\u27s reward. Using the theoretical results developed for the two problems, we propose an online hyperparameter optimizer called Hyperparameter Controller (HyperController) in Reinforcement Learning (RL) to improve the efficiency and performance of training RL neural networks. Our theoretical results demonstrate that HyperController accelerates the training phase while also consistently improving the neural network\u27s performance
Optical Imaging of Cortical Neurophysiology in Healthy Mice and Following Stroke
No full textMy thesis has centered on developing innovative optical imaging technology for 1) understanding neurophysiological processes in the brain and 2) endogenous mechanisms of plasticity after stroke. Towards the first goal, local changes in blood oxygenation and brain energy utilization are coupled to changes in neuronal activity (neurovascular coupling(NVC) and neurometabolic coupling(NMC)). However, a cohesive link between how local changes in neuronal activity manifest as changes in metabolic and hemodynamic signaling has not been established, in part due to technological limitations. I leveraged the red-shifted genetically-encoded calcium indicator, jRGECO1a, along with dual-fluorophore wide-field optical imaging(2F-WFOI) of spontaneous calcium, flavoprotein autofluorescence (as a measure of oxidative metabolism) and blood oxygenation in mice. I demonstrate regional differences in NVC and NMC in the awake mouse brain, and that anesthesia profoundly altered NVC but not NMC. Towards the second goal, parvalbumin inhibitory interneurons(PV-INs) are known to mediate several forms of activity-dependent plasticity. However, it is unknown whether PV-INs mediate brain repair processes after stroke. I created a novel imaging system for all-optical neuronal probing via optogenetic targeting of PV-INs and mesoscopic imaging of excitatory activity via jRGECO1a. Photostimulation over a grid in the left hemisphere allowed for visualizing a whole-cortex “PV connectome”. In healthy mice, PV-INs extend ipsilateral inhibitory influences spanning several mm and region-dependent interhemispheric inhibition. Photothrombosis of left forepaw cortex resulted in decreased effective connectivity cortex-wide at week 1. Recovery following stroke was associated with global reestablishment of PV-based inhibition. Future work will determine a causal role of PV-INs on recovery
Beyond Money: Early Asset-Building Policy Supports Mothers, Strengthens Families, and Builds Children’s Futures
No full textTo foster stronger futures for children and families in the United States, policymakers and researchers are increasingly turning their attention to early-life asset-building policies, recognizing that such policies are vital components of family well-being and support. In this brief, CSD researchers discuss evidence from the SEED for Oklahoma Kids experiment (SEED OK), which shows that Child Development Account policy shapes mothers, families, and children in profound ways. Child Development Accounts, they conclude, represent a scalable and bipartisan family-support policy, offering early and automatic foundation that reinforces parental hopes and planning while promoting both equity, through progressively structured design, and responsibility, through asset-building principles
International Space Law under the Trump Administration: Commerce and Security
No full textEarly themes of the second Trump administration\u27s orientation toward international space law include military assertiveness, privatization, and nominal diplomatic engagement. These themes carry significant implications for international space law. They pressure longstanding international legal ambiguities, widen existing geopolitical divisions, and raise questions about how public-private interdependence will shape state responsibility and the development of norms in space. This essay was prepared for a special issue of the American Journal of International Law that offers an introduction and overview of the U.S. government\u27s actions across a wide range of international legal issues in the first six months of President Trump\u27s second term
Hardships Increase as Households Rely More on Gig Work
No full textGig work has grown substantially in the United States in recent years. Although research has examined the relationship between participation in the gig economy and hardship, few studies examine how households’ reliance on gig work relates to hardship experiences. Drawing upon data from the Workforce Economic Inclusion and Mobility survey of a nationally representative sample of low-wage workers, this brief explores the relationship between the proportion of a household’s income from gig work and indicators of financial distress: bill payment difficulties, food insufficiency, budget constraints, blood plasma sales, reliance on alternative financial services such as payday loans, and usage of buy-now-pay-later loans
The Practice of Executive Constitutionalism
No full textThe Executive Branch must inevitably interpret the Constitution. Although departmentalists and judicial supremacists disagree about the scope of the Executive’s constitutional authority, few believe the Constitution is only for the courts. But what are the practices through which the Executive Branch interprets the Constitution and translates those interpretations into concrete decisions? What is their history? And what, if anything, is distinctive about them? While a rich and growing literature has examined some aspects of these questions, scholars have not broadly canvased the most central tools by which the Executive Branch shapes and implements constitutional law, or considered what makes them unique.
This Article pursues that project. Descriptively, the Article provides a thick account of Executive Branch constitutional interpretation, particularly in its centralized form controlled by the President and the Department of Justice. We describe and assess executive tools and methods for interpreting the Constitution and transmitting those interpretations to different audiences. Some of these tools are well known and have obvious judicial analogues. But the Article shows how the history and contours of these practices have not been fully understood. It also excavates some unfamiliar tools that have gone unnoticed and unexplained.
Our descriptive account provides a foundation for assessing executive constitutionalism. Comparing executive and judicial practices can help justify some existing arrangements while suggesting reforms for others. More broadly, a rich understanding of how Executive Branch constitutional interpretation has worked is critical for assessing the virtues and vices of executive constitutionalism writ large—especially in the second Trump Administration, in which expansive claims of constitutional authority loom large
Constitutional limitations and the rule of law: an Indian perspective
No full textThis dissertation undertakes a comprehensive examination of the doctrine of the rule of law within the Indian constitutional framework, with particular emphasis on the judiciary’s role in interpreting constitutional limitations on legislative and executive power. Since the enforcement of the Constitution of India in 1950, the Indian Supreme Court has played a pivotal role in defining, operationalizing, and enforcing the rule of law through judicial review. However, this role has been accompanied by persistent doctrinal uncertainty, especially regarding the appropriate methodology of constitutional interpretation. This study critically explores how competing interpretative approaches—namely living constitutionalism, originalism, and the doctrine of basic structure—have shaped Indian constitutional jurisprudence and assesses their compatibility with the foundational ideals of the rule of law
Analyzing Sleep Staging and Cognitive Transitions Using fMRl Time Series Analysis
No full textUnderstanding sleep stages and their underlying cognitive transitions is essential for advancing our knowledge of brain function. While sleep staging has traditionally relied on electroencephalogram (EEG), this study explores the feasibility of classifying sleep stages using only functional Magnetic Resonance Imaging (fMRI) data. We investigate the use of Hidden Markov Models (HMM) and Hierarchical Dirichlet Process Hidden Markov Models (HDP-HMM) to capture temporal brain-state dynamics from task-free fMRI recordings. A series of pipelines—including sliding window, two-stage majority filtering, and prototype-based mapping via the Hungarian algorithm—were implemented to align model-inferred states with EEG-defined sleep stages. Among the twelve tested pipelines, the HMM with majority filtering and prototype mapping achieved the highest overall classification accuracy while maintaining interpretability and computational efficiency. Analyses of transition probabilities, dwell times, and fractional occupancy revealed that the HMM-derived states exhibited weak correspondence with EEG sleep stages in both temporal structure and functional connectivity patterns. Limitations emerged due to the low temporal resolution and high dimensionality of fMRI data, which hindered the detection of short-lived or rare sleep stages such as REM. These results demonstrate the potential of unsupervised HMM-based approaches for fMRI-only sleep staging and highlight the need for future enhancements through semi-Markov models, adaptive smoothing, and integration of dimensionality reduction techniques to better capture transient cognitive states
Parametric Control and Amplification in Superconducting Quantum Systems: From Entanglement Probes to PT-Symmetric Architectures
No full textThe precise control, entanglement, and measurement of quantum systems are central challenges in realizing scalable quantum information processing. This dissertation investigates how parametric modulation and engineered nonlinearity can be harnessed to both manipulate and measure superconducting qubits with high fidelity. In one part, controlled parametric drives are used to generate entanglement between qubit pairs and employ their joint states as probes of environmental dynamics. This approach enables the exploration of the non-Markovian nature of quantum environments, revealing how memory effects influence qubit coherence and energy relaxation. In a separate but conceptually related study, the same principle of parametric driving is utilized to design and analyze quantum-limited amplifiers based on three-wave mixing in weakly nonlinear Josephson circuits. An amplifier with a higher coupling quality factor than conventional JPAs is employed, allowing the two modes of a parity-time (PT) symmetric dimer to be mapped onto the in-phase (I) and quadrature (Q) components of the Josephson parametric amplifier. This mapping enables controlled exploration of non-Hermitian dynamics and mode coalescence within the amplifier’s quadrature space, offering new insights into PT-symmetric behavior and quantum-limited detection. Collectively, these investigations unify the themes of parametric control, entanglement, and amplification, advancing the broader understanding of quantum measurement and open-system dynamics
Administrative Law and Process in a Nutshell, 7th Edition
No full textThis book offers a concise, knowledgeable guide to administrative law. In straightforward, readable prose, the authors not only summarize the dominant statutes and case law in the area, but also discuss informal administrative processes and the background realities of the regulatory state. Students can use the book as a complement to any major casebook, and practitioners will also find it an excellent brief introduction to this complex and important subject