12023 research outputs found
Sort by
The Woman Who Waits No More
No full text[Introduction] Homer’s epic poems are stories written mostly about men and the intricacies of their roles
in Ancient Greece, choosing to focus on grief and masculinity and exploring the consequences of
hubris and plays for power. As a result, the stories about the inner lives of the women of ancient
epic are often overshadowed by the roles they play in forwarding and motivating the narratives
of men. One such story is the narrative of Penelope, the faithful, ever-waiting wife of Odysseus.
The accounts of her longing for her husband and the peril she and her household face at the
hands of the suitors are largely responsible for creating the sense of urgency for Odysseus’s
return home which underpins his misadventures at sea. But the Odyssey lacks a fuller
exploration of who Penelope became over the twenty long years of Odysseus’s absence during
which time she raised a son, ruled over a nation, and weathered the onslaught of the suitors.
Though outside of the scope of the Odyssey, Carol Ann Duffy and A. E. Stallings in their poems
“Penelope” and “The Wife of the Man of Many Wiles” are able to explore Penelope’s character
in more depth, especially regarding her portrayal as a woman who is passively waiting for
change. Duffy’s poem develops Penelope as a character who decides that she would rather take
control of her narrative than spend her days waiting upon a husband who may never return. The
focus of the poem centers on her development largely during the time before the arrival of the
suitors, and when they come, focuses on her ability to deceive them without masculine aid – a
narrative which is similar to the one which already exists within the Odyssey. On the other hand,
in “The Wife of the Man of Many Wiles”, Stallings focuses directly on Penelope’s battle with the
suitors, criticizing her perception as a woman who waits by implying that there was no stalemate
for Odysseus to break between the desiring suitors and the “faithful” Penelope after all, that by
the time he arrives he is too late to save her from being undone by the suitors, no matter what he
chooses to believe
From Source to Sink: Measuring and Modeling Processes Affecting Methane Emissions and Loss
Get PDFMethane is a key target for climate change mitigation efforts. With a radiative forcing 85 times stronger than CO₂ over a 20-year period and an atmospheric lifespan of only a decade, mitigating methane emissions will slow climate change in the near-term. However, quantifying methane emissions from specific sectors accurately poses a significant challenge. This is because top-down estimations of methane emissions demand precise observations and constraints on a range of physical and chemical processes. In this thesis, I seek to enhance the accuracy of methane emissions calculations by resolving these processes in detail and advocating for an expansion of the methane monitoring network.
The primary mechanism for atmospheric methane destruction is its oxidation by the Hydroxyl radical (OH). Chemical feedbacks due to temporal variations in OH availability can substantially influence the methane lifetime and, consequently, emissions trends over recent decades. In Chapter 2, I quantify the impact of this predominant chemical loss mechanism on methane emissions calculations.
Methane loss to the stratosphere represents the second most significant methane destruction mechanism, although the processes involved remain highly uncertain. Accurately quantifying methane loss via stratospheric-tropospheric exchange is crucial for improving the accuracy of methane emissions calculations. In Chapter 3, I utilize chemical tracers to determine how stratospheric-tropospheric exchange influences global methane emissions trends.
Current understanding of greenhouse gas fluxes from a top-down perspective typically relies on atmospheric inversions, which depend on spatial and temporal gradients in observed greenhouse gas concentrations. However, maintaining highly accurate ground-based measurements poses logistical and financial challenges, while satellites currently do not provide the requisite accuracy and spatial resolution for long-term monitoring. In Chapter 4, I explore the potential of frequency combs in measuring environmental impacts on greenhouse gas sensing and as tools to expand the observation network.
In summary, this thesis contributes to a more profound understanding of the two primary methane sinks and how their variations affect methane emissions trends over recent decades. It also lays the groundwork for the next-generation greenhouse gas observation network using laser frequency combs by quantifying environmental impacts on greenhouse gas spectroscopy directly in the field. Future advances should focus on a more accurate understanding of methane sink processes, improved spectroscopy, and expanded measurement networks. This will require advances in both modeling and measurements.
Ultimately, rapid and efficient mitigation of methane emissions remains the most feasible approach to curb anthropogenic climate change. To do this however, accurate assessments of methane trends and emissions necessitate bringing methane measurements and modeling of methane destruction processes closer to the real world.</p
New to Nature C–C Bond Forming Cyclases: Pushing the Boundaries of Ring Forming Reactions
Get PDFBiocatalysts have shown themselves to be extremely powerful for the synthesis of pharmaceuticals, fragrances, and fine chemicals, providing products with high yields and selectivities. Recently, new-to-nature biocatalysis has received increased attention, allowing for the benefits of biocatalysis to be applied to reactions that were previously the sole domain of chemocatalysts. Engineers have begun to develop enzymes that catalyze new-to-nature C–C bond forming cyclisation reactions, which are quite powerful due to their ability to build the carbon skeleton of molecules. Despite this, this class of enzymes is limited in scope. This thesis details the expansion of C–C bond forming cyclases, including expanding the scope of cytochrome P411 cyclopropanation and an intramolecular C–H functionalization strategy for the synthesis of diverse rings. Chapter 1 introduces biocatalysis and its recent applications, especially as they apply to new-to-nature C–C bond forming cyclisation reactions. Chapter 2 shows the development of a cytochrome P411 that catalyzes the enantio- and diastero-specific synthesis of 1,2,3-polysubstituted cyclopropanes. Using directed evolution, this carbene transferase was evolved to react with internal alkenes and build two C–C bonds, expanding the scope and specificity of cyclopropanation reactions. Chapter 3 describes the expansion of this biocatalytic system toward the synthesis of stereoconvergent products, enabling more efficient synthesis from non-diasteropure starting materials. Chapter 4 details the evolution of a cytochrome P411 to perform an intramolecular C–H functionalization using diazo compounds, making a variety of differently sized rings with different molecular geometries. In summary, this work addresses the need for expansion of new-to-nature C–C bond forming cyclisation reactions and provides a guide for expanding new-to-nature reactions to their full potential.</p
AI for Scientists: Accelerating Discovery Through Knowledge, Data, and Learning
Get PDFWith rapidly growing amounts of experimental data, machine learning is increasingly crucial for automating scientific data analysis. However, many real-world workflows demand expert-in-the-loop attention and require models that not only interface with data, but also with experts and domain knowledge. My research develops full stack solutions that enable scientists to scalably extract insights from diverse and messy experimental data with minimal supervision. My approaches learn from both data and expert knowledge, while exploiting the right level of domain knowledge for generalization. This thesis presents progress towards developing automated scientist-in-the-loop solutions, including methods that automatically discover meaningful structure from data such as self-supervised keypoints from videos of diverse behaving organisms. We will then discuss methods that use these interpretable structures to inject domain knowledge into the learning process, such as guiding representation learning using symbolic programs of behavioral features computed from keypoints. This work is the result of close collaborations with domain experts, such as behavioral neuroscientists, in order to identify bottlenecks and integrate these methods in real-world workflows. My aim is to enable AI that collaborates with scientists to accelerate the scientific process
How to Make Small Things Do Big Things: Exploring Engineered Disorder for Massively Scalable Metasurfaces and Metamaterials
Get PDFThis work presents a collection of topics related to anomalous electromagnetic scattering, emission, and absorption states formed from random systems. The underlying motivation is to explore to what extent metasurface and metamaterial concepts could be applied at a massively large scale; by identifying emergent properties in systems that do not require careful fabrication. Emphasis is placed on exploring theoretical descriptions for systems that do not conform well to existing simpler models. Covered topics include random metasurfaces for spectral filtering and polarization invariance, random nanoparticle films for radiative cooling, broadband polarization and angle invariant absorption using random fractals, effective medium models beyond traditional assumptions, a mathematical transform to understand highly directional scattering/emission in complex systems, and optical metrology and characterization techniques for random systems
Linear Amplification in Nonequilibrium Turbulent Boundary Layers
Get PDFResolvent analysis is applied to nonequilibrium incompressible adverse pressure gradient (APG) turbulent boundary layers (TBL) and hypersonic boundary layers with high temperature real gas effects, including chemical nonequilibrium. Resolvent analysis is an equation-based, scale-dependent decomposition of the Navier Stokes equations, linearized about a known mean flow field. The decomposition identifies the optimal response and forcing modes, ranked by their linear amplification. To treat the nonequilibrium APG TBL, a biglobal resolvent analysis approach is used to account for the streamwise and wall-normal inhomogeneities in the streamwise developing flow. For the hypersonic boundary layer in chemical nonequilibrium, the resolvent analysis is constructed using a parallel flow assumption, incorporating N₂, O₂, NO, N, and O as a mixture of chemically reacting gases.
Biglobal resolvent analysis is first applied to the zero pressure gradient (ZPG) TBL. Scaling relationships are determined for the spanwise wavenumber and temporal frequency that admit self-similar resolvent modes in the inner layer, mesolayer, and outer layer regions of the ZPG TBL. The APG effects on the inner scaling of the biglobal modes are shown to diminish as their self-similarity improves with increased Reynolds number. An increase in APG strength is shown to increase the linear amplification of the large-scale biglobal modes in the outer region, similar to the energization of large scale modes observed in simulation. The linear amplification of these modes grows linearly with the APG history, measured as the streamwise averaged APG strength, and relates to a novel pressure-based velocity scale.
Resolvent analysis is then used to identify the length scales most affected by the high-temperature gas effects in hypersonic TBLs. It is shown that the high-temperature gas effects primarily affect modes localized near the peak mean temperature. Due to the chemical nonequilibrium effects, the modes can be linearly amplified through changes in chemical concentration, which have non-negligible effects on the higher order modes. Correlations in the components of the small-scale resolvent modes agree qualitatively with similar correlations in simulation data.
Finally, efficient strategies for resolvent analysis are presented. These include an algorithm to autonomously sample the large amplification regions using a Bayesian Optimization-like approach and a projection-based method to approximate resolvent analysis through a reduced eigenvalue problem, derived from calculus of variations.</p
Thermal Kinetic Inductance Detector (TKIDs) Camera: A Pathfinder mm-Wave Polarimeter
Get PDFThermal kinetic inductance detectors (TKIDs) are novel, superconductive, frequency-multiplexed bolometric detectors that promise the same noise performance as traditional transition-edge superconducting bolometers, but with ease of scalability to large array formats. This research starts with TKIDs being in the early stage of development. Readout hardware and strategies were developed to characterize the first batches of devices, still in the chips form containing a few detectors. The success of the TKIDs chips lead to a more in-depth exploration of the detector physics and the possibility to scale from chips containing tens of detectors to tiles counting hundreds. Novel characterization techniques were developed to deal with the testing of the arrays in a laboratory environment. Finally, the characterization of a science grade tile is detailed
Reputation and Accountability
Get PDFIn this thesis, I explore how accountability relationships affect policymaking in two institutional contexts: internal executive branch operations and electoral contests. The overarching insight is that the potential for removal creates reputation concerns to demonstrate skill that, in turn, affect policymaking. For political appointees serving at the pleasure of the president, this means a reputation for management skill or technocratic policy expertise, whereas for elected representatives, this means maintaining a reputation for competent leadership with voters. The main result is that oversight creates both pathological policymaking incentives for accountable officials, but also potentially unintuitive selection by a principal—either the president or voters.
In Chapters 1 and 2, I explore political appointees’ dual roles as agents of the president and managers of the bureaucracy. This view of appointee-careerist relations complicates standard notions of presidential control and bureaucratic power, by recognizing that appointees are reliant on presidential support to maintain their position within an administration. To cultivate a good reputation with the president, appointees may cede control to the bureaucracy. However, to understand how control is transferred to the bureaucracy, I argue that we must fully account for appointees’ strategic roles in the administrative presidency—and that, to do so, requires differentiating between types of appointments.
Presidential appointments that require Senate confirmation (PAS) and noncareer members of the Senior Executive Service (SES-NA) occupy positions that require direct oversight and management of subordinate career civil servants. As managers, these appointees must rely on the expertise, pragmatic or otherwise, and efforts of bureaucrats to implement the president’s policies. I argue that presidents select these appointees primarily on the basis of their management skills. In contrast, Schedule C appointees occupy confidential or policymaking roles and serve directly under a political appointee. These appointees may substitute for the expertise of career bureaucrats. I argue that presidents select these appointees on the basis of policy expertise.
However, central to my argument is the idea that the president may still be uncertain of an appointee’s management skill or policy expertise—despite appointing him or her in the first place. This means there is scope for the president to learn about an appointee’s ability based on how they perform or behave on the job. It is this residual uncertainty about an appointee’s capabilities, along with the president’s formal removal power, that create reputation concerns for appointees: appointees care about maintaining their position and to do so they must preserve their reputation with the president.
I argue that these reputation concerns shape how appointees manage interactions with the bureaucracy. Appointees in managerial roles may make more policy concessions to the bureaucracy than the president would like in order to ensure bureaucratic cooperation and avoid revealing managerial weaknesses. Instead, appointees in positions of policymaking authority may fail to empower or involve bureaucrats in policymaking. Both of these actions undermine the president’s policy goals by either creating policies that increasingly reflect the views of the bureaucracy or by failing to create policies that reflect bureaucratic expertise. This suggests limitations of political control over the bureaucracy that cannot be alleviated through the exercise of formal administrative powers, namely appointment and removal powers.
Ultimately, the agency issues I explore in this context follow from a fundamental and immutable constraint on presidential control: the president simply cannot unilaterally manage the executive branch. The demands of the presidency are too great for the president to preside over all operations. This means delegation is necessary—and, even when the president delegates to advisors of “her own choosing,” some loss of control is inevitable.
In Chapter 3, I explore how majority selection operates in an environment in which politicians prefer to pursue particularistic policies. If special interest coalitions are sufficiently strong, a majority may expect that political expertise will be used to select policies that generate rents for narrow constituencies at the expense of its own welfare. I develop a model in which a majority prefers to elect the less competent politician in order to undermine the incumbent’s ability to pursue the special interest agenda and derive the implications for accountability in this setting. The results demonstrate that the majority’s attempts to reassert control over policy through its retention decisions impede social welfare maximizing reform and distort aggregate welfare by either encouraging (i) inefficient policy selection or (ii) inefficient candidate selection. Even if politicians choose policies that maximize social welfare doing so may only worsen aggregate welfare by providing voters with more information about candidate competence, which enables the majority to better select inept politicians.</p
Early Dynamics and Evolution of Extrasolar Planetary Systems
Get PDFOf the thousands of discovered exoplanets, the vast majority were born billions of years ago. The process of their formation was only a tiny fraction of their lifespan and observing formation of new planets is very difficult with current techniques. However, these planets, and the planetary systems they are a part of, retain distinct fingerprints of how and when they formed. This thesis presents six studies that aim to uncover the environment in which planets form by investigating how the architectures of multiplanet systems are shaped by physical processes. I show that varying degrees of planet-planet interactions, planet-disk interactions, and tidal dissipation successfully reproduce many bulk features of the small planet census. Furthermore, analysis of selected individual systems can recover detailed measurements of the protoplanetary disk environment and orbital histories of the planets. Similar processes unfold in the satellite systems of giant planets, which are akin to scaled-down exoplanet systems.</p
Data-Driven Safety-Critical Autonomy in Unknown, Unstructured, and Dynamic Environments
Get PDFThis thesis addresses the critical challenge of ensuring safety in autonomous exploration within unknown, unstructured, dynamic environments, a domain filled with various types of uncertainties. These include model uncertainties in system dynamics, localization uncertainties stemming from measurement noises, and the risks of collision in environments with dynamic obstacles. Traditional models for vehicle planning and control are often simplified for computational feasibility, but this simplification without careful analysis can compromise safety and system stability. My research introduces a novel, comprehensive framework to provide probabilistically safe planning and control for robot autonomy, structured around three components:
(1) Probabilistic Uncertainty Quantification for Model Mismatches:
This segment focuses on identifying model discrepancies given closed-loop tracking data in an unstructured environment where a reduced-order robot model is used for planning and control. The disturbance is modeled as a scalar-valued stochastic process of a norm on the difference between the reduce-order robot model and actual system evolution. In an online and risk-aware framework, Gaussian Process Regression is employed to extract the probabilistic upper bound to such stochastic process, referred to as the Surface-at-Risk. Theoretical guarantees on the accuracy of the fitted discrepancy surface are analyzed and verified to the data sets collected during system operation.
In an offline setting, conformal prediction, a statistical inference tool, is employed to obtain probabilistic upper bounds of matched and unmatched model disturbance in the system from data, without any assumption of the latent probability distribution governing these discrepancies. Building on these bounds, the robot's nominal ancillary controller is augmented for extending robustness and stability guarantees of the closed-loop system in the face of such discrepancies. Additionally, a maximum tracking error tube is constructed along the planned trajectory using the reduced-order model. Such error tubes describe the maximum permissible deviation in actual trajectory tracking under the augmented ancillary controller and the worst-case matched and unmatched model uncertainties, thereby delineating safe operational boundaries for the system.
(2) Data-Driven Unsafe Set Prediction for Dynamic Obstacles:
This thesis topic develops an online, data-driven predictive model for dynamic obstacles, accounting for measurement noise and low-frequency data rates.
First inspired by singular spectrum analysis (SSA), a time-series forecast technique, obstacle models characterized by linear recurrence relationships are extracted from real-time position observables. Using the statistical bootstrap technique, a set of predicted obstacle trajectories are constructed, which in turn are reformulated into deterministic distributionally robust obstacle avoidance constraints, reflecting a user-defined risk tolerance.
Further refining the obstacle predictor for intention-unknown obstacles, a linear, time-varying model is learned from data using time-delay embedding of obstacle position observables. Additive process and measurement noises are anticipated in the learned model, where their intensities are estimated from data. For inferring prediction uncertainties, a companion data-driven Kalman Filter (DDKF) is constructed to forecast obstacle positions and uncertainties. This "heuristic unsafe set" from DDKF is then dynamically calibrated using adaptive conformal prediction, ensuring safety without relying on any distribution assumptions regarding the uncertainties or model accuracy. The calibrated sets, called conformal prediction sets, are then reformulated into convex state constraints.
(3) Safety-Critical Planning:
The thesis proposes two methods for ensuring safety in planning and navigation: Probabilistic-Safe Model Predictive Control (MPC) and Probabilistic-Safe Model Predictive Path Integral (MPPI) given uncertainties arising from operating in unknown, unstructured, and dynamic environments. The MPC approach integrates the quantified obstacle avoidance constraints into a convex program to balance computational tractability while providing probabilistic safety guarantees. In contrast, the MPPI method, a sampling-based strategy, incorporating unsafe sets into a cost map derived from sensory data, optimizes reference tracking trajectory while guaranteeing collision avoidance up to a user-defined risk tolerance.
In unknown and cluttered environments automatically, the proposed framework learns an upper bound on model residuals from data and systematically calculates the safety buffers needed to provide the desired probabilistic safe navigation of robotics systems. Additionally, in the presence of dynamic obstacles, the proposed data-driven predictor systematically extracts an obstacle model and makes obstacle-occupied unsafe set forecasts. These features largely eliminate the "hand tuning" of the underlying planner and controller that is normally required in heuristic-based algorithms. The efficacy of these proposed frameworks is empirically validated through Monte Carlo Simulations, alongside hardware validations on both ground and aerial vehicles, demonstrating their robustness, versatility, and applicability in real-world scenarios.</p