150813 research outputs found
Sort by
t-channel dark matter models – a whitepaper
No full textThis report, summarising work achieved in the context of the LHC Dark Matter Working Group, investigates the phenomenology of t-channel dark matter models, spanning minimal setups with a single dark matter candidate and mediator to more complex constructions closer to UV-complete models. For each considered class of models, we examine collider, cosmological and astrophysical implications. In addition, we explore scenarios with either promptly decaying or long-lived particles, as well as featuring diverse dark matter production mechanisms in the early universe. By providing a unified analysis framework, numerical tools and guidelines, this work aims to support future experimental and theoretical efforts in exploring t-channel dark matter models at colliders and in cosmology
Cultivating a Supportive Sphere: Designing Technology to Increase Social Support for Foster-Involved Youth
No full textApproximately 400,000 youth in the US are living in foster care due to experiences with abuse or neglect at
home [17]. For multiple reasons, these youth often don’t receive adequate social support from those around
them. Despite technology’s potential, very little work has explored how these tools can provide more support
to foster-involved youth. To begin to fill this gap, we worked with current and former foster-involved youth
to develop the first digital tool that aims to increase social support for this population, creating a novel system
in which users complete reflective check-ins in an online community setting. We then conducted a pilot study
with 15 current and former foster-involved youth, comparing the effect of using the app for two weeks to
two weeks of no intervention. We collected qualitative and quantitative data, which demonstrated that this
type of interface can provide youth with types of social support that are often not provided by foster care
services and other digital interventions. The paper details the motivation behind the app, the trauma-informed
design process, and insights gained from this initial evaluation study. Finally, the paper concludes with
recommendations for designing digital tools that effectively provide social support to foster-involved youth
Blueprints for the Geometric Control of N-Heterocyclic Carbene–Carbodiimide Isomers
No full textRational control of the 3D presentation of atoms—stereochemistry—lies at the heart of synthetic organic and materials chemistries. Here, researchers report detailed computational studies on conformational isomerism in N-heterocyclic carbene–carbodiimide (NHC–CDI) zwitterionic adducts. By varying the steric and electronic parameters of the NHC and CDI components, criteria for controlling isomerization thermodynamics and predicting energetically favorable conformations are identified. These criteria is validated experimentally using a novel synthetic approach to NHC–CDIs, which exploits the thermodynamic equilibrium between sterically unencumbered NHC dimers to access NHC–CDI adducts with low barriers to conformational isomerization, including the first example of an (E/E)-NHC–CDI
A Comparison of Theoretical and Actual Coumarin Exudation Under Iron Limitation to Understand Root Exudation Mechanics
No full textNutrient cycling is an important component of plants’ immune systems, largely driven by the act of exuding environmentally influential metabolites from roots. Root exudation may be driven by multiple unique mass-transport mechanisms, including active and passive transport types, though the latter is not well-studied despite being labelled a significant driver of low molecular weight metabolite exudation. This research investigates the generally accepted assumption that low molecular weight metabolites, including iron-fixing coumarins (scopoletin, fraxetin, etc.) are primarily exuded passively, and high molecular weight metabolites follow an active exudation approach. Scopoletin and scopolin exudation from Arabidopsis thaliana in low-iron and replete conditions is quantified to determine if the hypothesized passive diffusion mechanism is a significant contributor to coumarin exudation. LC-MS analysis suggests that passive diffusion of scopoletin and scopolin from roots plays a significant role in total coumarin exudation values. Further research should include investigating the implications of passive coumarin exudation on long-term iron storage and soil health in addition to the relationship between coumarin production and exudation.M.Eng
Mitigating LLM Hallucination in the Banking Domain
No full textLarge Language Models (LLMs) offer significant potential in the banking sector, particularly for applications such as fraud detection, credit approval, and enhancing customer experience. However, their tendency to "hallucinate"—generating plausible but inaccurate information—poses a critical challenge. This thesis examines existing strategies for mitigating LLM hallucinations and proposes a novel approach to reduce hallucinations in the context of predicting customer churn using LLMs.M.Eng
Adaptive Wavefront Estimation Algorithms for High-Contrast Imaging of Exoplanets
No full textThe direct imaging of exoplanets orbiting stars outside our solar system remains one of the crucial tools we have available to answer whether there exists life beyond Earth. The light from an Earth-like exoplanet is approximately ten orders of magnitude dimmer than its host star and hence the imaging system of the telescope observing the exoplanet must be able to suppress the starlight to achieve a “contrast” of 10−10 in the image. This is typically achieved using a coronagraph, which blocks the light from the star while allowing the light from the planet to pass through. However, some starlight that leaks through the coronagraph needs to be further removed in the search region for the exoplanet; this region is referred to as the dark hole or dark zone (DZ). Creating a DZ requires the use of focal plane wavefront sensing and control techniques, which estimates the electric field of the starlight in the focal plane of the telescope using a camera and then informs the deformable mirrors (DMs) located upstream of the coronagraph to null these electric fields. Once the DZ is created with a desired contrast, there are still slow, high-order drifts in the optical system that cause the contrast to degrade over the long observation times of the science target. High-order wavefront sensing and control (HOWFSC) techniques are required to maintain the contrast in the DZ while observing a science target. Dark Zone Maintenance (DZM) is a technique that has demonstrated the ability to maintain the contrast in the DZ over long observation times. This algorithm utilizes an Extended Kalman Filter (EKF) to estimate the open-loop electric field at every pixel in the DZ and use this information to inform the control algorithm. The achievable contrast and contrast stability of DZM are determined by several key parameters: the optical system’s drift rate, the photon flux and associated shot noise in the measurement images, and the probe magnitude applied to the DMs for the estimation algorithm. This work quantifies the impact of the drift rate, photon rate, and probe magnitude on the performance of DZM by performing a parameter scan on high-contrast imaging testbeds. The parameter scan was performed on both the in-air High-contrast imager for Complex Aperture Telescopes (HiCAT) testbed at the Space Telescope Science Institute (STScI) and the in-vacuum Decadal Survey Testbed (DST) at the Jet Propulsion Laboratory (JPL). The parameter scan was run in both simulation and on the physical testbed using the contrast in the DZ as a performance metric, and evaluated relative to the photon-noise theoretical bounds to assess the efficacy of the DZM algorithm. The substantial difference between the theoretical bounds and experimental results, on average 70 times worse on HiCAT, motivated the development and implementation of a new DZM algorithm that utilized a separate EKF to estimate the modes of wavefront error derived from the DMs and use that information to correct for the aberrations. This new modal EKF algorithm was tested with a similar parameter scan on the HiCAT simulator demonstrating a nearly 5 times level of improvement relative to the original DZM algorithm simulation performance. The results of this work will inform the design of future algorithms to maintain high contrast during observations for upcoming space telescope missions such as the Habitable Worlds Observatory (HWO).S.M
Biosensor development for single-cell detection of glucuronate
No full textRecent work in biosensors has shown promise to enable high throughput searches through large genetic libraries. However, just as physiological limitations and lack of in-depth mechanistic knowledge can prevent us from achieving high titers in microbial systems; similar roadblocks can appear in the application of biosensors. Here, we characterized a previously developed transcription-factor (ExuR) based galacturonate biosensor for its other cognate ligand, glucuronate. Though we saw an ideal response to glucuronate from the biosensor in controlled and ideal experimental circumstances, these results began to deviate from a well-behaved system when we explored the application of the sensor to different MIOX homologs. Through modifications to circuit architecture and culture conditions, we were able to decrease this variation and use these more optimal conditions to apply the biosensor for the separation of two closely related MIOX homologs
FPGA Based Data Acquisition System for Cryogenic Device Verification
No full textIn this work, a system of processors connected to an FPGA is interfaced with a custom analog frontend and used to create a verification environment for cryogenic devices. In particular, this thesis focuses on the technical structure of that system. Current validation efforts often rely on commercially available arbitrary waveform generators (AWGs) and oscilloscopes, which, while highly capable, are often prohibitively expensive and poorly suited for large-scale or parallelized testing environments. As noted in industry reports, scaling such instrumentation introduces significant challenges in cost, calibration, and signal synchronization, making them inefficient for high-resolution or high-speed analyses in multi-channel systems [1]. On the other hand, an FPGA provides the necessary performance to increase parallelism without a proportional increase in cost, greatly improving testing resolution and speed. When augmented with a set of processors, we introduce a level of accessibility and automatability not currently present in commercial products. To be clear, while the board was designed with the testing of nanowires in mind (and is not capable of measuring DC voltages), it can still be combined with separate lab equipment to interact with Josephson Junction based devices. That said, the flexibility of this system allows for a generalized application to any electronic that demands a specialized testing procedure involving arbitrary signal processing and generation. The money, time, and energy that this innovation will save on cryogenic electronic validation will significantly improve our progress in developing these technologies.M.Eng
Report to the President for year ended June 30, 2025, Department of Urban Studies and Planning
No full textThis report contains the following sections: Promotions and Faculty Appointments; Comings, Goings, Changing Roles; Committees and Leadership; Major Awards, Events, and Other Noteworthy News; Education/Degree Programs; Commencement/Awards; and DUSP Student Council Awards
Sampling-based sublinear low-rank matrix arithmetic framework for dequantizing quantum machine learning
No full textWe present an algorithmic framework for quantum-inspired classical algorithms on close-to-low-rank matrices, generalizing the series of results started by Tang’s breakthrough quantum-inspired algorithm for recommendation systems [STOC’19]. Motivated by quantum linear algebra algorithms and the quantum singular value transformation (SVT) framework of Gilyén et al. [STOC’19], we develop classical algorithms for SVT that run in time independent of input dimension, under suitable quantum-inspired sampling assumptions. Our results give compelling evidence that in the corresponding QRAM data structure input model, quantum SVT does not yield exponential quantum speedups. Since the quantum SVT framework generalizes essentially all known techniques for quantum linear algebra, our results, combined with sampling lemmas from previous work, suffice to generalize all prior results about dequantizing quantum machine learning algorithms. In particular, our classical SVT framework recovers and often improves the dequantization results on recommendation systems, principal component analysis, supervised clustering, support vector machines, low-rank regression, and semidefinite program solving. We also give additional dequantization results on low-rank Hamiltonian simulation and discriminant analysis. Our improvements come from identifying the key feature of the quantum-inspired input model that is at the core of all prior quantum-inspired results: ℓ2-norm sampling can approximate matrix products in time independent of their dimension. We reduce all our main results to this fact, making our exposition concise, self-contained, and intuitive