arXiv subjects

Scott Fleming

Publications and source records attributed to Scott Fleming.

8 recordsLinked to original sources

arXiv2026

The Hubble Advanced Spectral Product (HASP) Program

The Hubble Advanced Spectral Products (HASP) program is designed to robustly coadd Cosmic Origins Spectrograph (COS) and Space Telescope Imaging Spectrograph (STIS) spectra within the Mikulski Archive for Space Telescopes (MAST) in an automated fashion such that coadds are available for new data or archival data with updated calibrations. For each target within a visit or program, HASP employs a meticulous multi-stage filtering process to ensure data quality and creates coadded products for all central wavelengths (CENWAVEs) within specific gratings, as well as combined products using different gratings and instruments. The project also emphasizes making the code accessible to the user community for custom coaddition. As calibrations improve and new data are added to the archive, HASP products are re-created automatically so that they represent the best reduction of a given visit or program. Automated coadditions like those achieved by HASP can significantly enhance the combination of different CENWAVES, increase signal-to-noise ratios, and increase wavelength coverage. These properties make HASP a vital resource for astronomers using archival spectroscopic data from HST.

astro-ph.IM↗

arXiv2026

Overview of the New Hubble Spectroscopic Legacy Archive

The new Hubble Spectroscopic Legacy Archive (HSLA) provides coadded spectra of individual targets that have been observed with the Cosmic Origins Spectrograph (COS) and the Space Telescope Imaging Spectrograph (STIS) over their operating lifetime. HSLA uses data available in the Mikulski Archive for Space Telescopes (MAST). It automatically produces coadds whenever new data become publicly available or when there is newly recalibrated data. HSLA defines individual targets by their associated coordinates, accounting for proper motions, and uses SIMBAD, NED and the Phase II observing proposals to obtain astronomical classifications for each object. Coadded spectra are produced for each observing mode. In the case of COS far-ultraviolet observations there is one coadded spectrum for each lifetime position (LP). Additionally, a spectrum spanning the entire wavelength range covered by the observations is produced by abutting the spectra from a selection of individual modes. For each individual target, HSLA also provides a human-readable metadata file with key information that can be used in searches or for further exploration of the data. The HSLA project also makes the code used for coadding spectra publicly available along with several other tools (using Jupyter notebooks) for custom coaddition required in special cases. In this report we will describe the main components of HSLA and provide a brief description of how the data and metadata can be accessed.

astro-ph.IM↗

arXiv2025

Reflections from Research Roundtables at the Conference on Health, Inference, and Learning (CHIL) 2025

The 6th Annual Conference on Health, Inference, and Learning (CHIL 2025), hosted by the Association for Health Learning and Inference (AHLI), was held in person on June 25-27, 2025, at the University of California, Berkeley, in Berkeley, California, USA. As part of this year's program, we hosted Research Roundtables to catalyze collaborative, small-group dialogue around critical, timely topics at the intersection of machine learning and healthcare. Each roundtable was moderated by a team of senior and junior chairs who fostered open exchange, intellectual curiosity, and inclusive engagement. The sessions emphasized rigorous discussion of key challenges, exploration of emerging opportunities, and collective ideation toward actionable directions in the field. In total, eight roundtables were held by 19 roundtable chairs on topics of "Explainability, Interpretability, and Transparency," "Uncertainty, Bias, and Fairness," "Causality," "Domain Adaptation," "Foundation Models," "Learning from Small Medical Data," "Multimodal Methods," and "Scalable, Translational Healthcare Solutions."

cs.LG↗

arXiv2023

The Shaky Foundations of Clinical Foundation Models: A Survey of Large Language Models and Foundation Models for EMRs

The successes of foundation models such as ChatGPT and AlphaFold have spurred significant interest in building similar models for electronic medical records (EMRs) to improve patient care and hospital operations. However, recent hype has obscured critical gaps in our understanding of these models' capabilities. We review over 80 foundation models trained on non-imaging EMR data (i.e. clinical text and/or structured data) and create a taxonomy delineating their architectures, training data, and potential use cases. We find that most models are trained on small, narrowly-scoped clinical datasets (e.g. MIMIC-III) or broad, public biomedical corpora (e.g. PubMed) and are evaluated on tasks that do not provide meaningful insights on their usefulness to health systems. In light of these findings, we propose an improved evaluation framework for measuring the benefits of clinical foundation models that is more closely grounded to metrics that matter in healthcare.

cs.LG↗

arXiv2022

Extending Optical Flare Models to the UV: Results from Comparing of TESS and GALEX Flare Observations For M Dwarfs

The ultraviolet (UV) emission of stellar flares may have a pivotal role in the habitability of rocky exoplanets around low-mass stars. Previous studies have used white-light observations to calibrate empirical models which describe the optical and UV flare emission. However, the accuracy of the UV predictions of models have previously not been tested. We combined TESS optical and GALEX UV observations to test the UV predictions of empirical flare models calibrated using optical flare rates of M stars. We find that the canonical 9000 K blackbody model used by flare studies underestimates the GALEX NUV energies of field age M stars by up to a factor of 6.5$\pm$0.7 and the GALEX FUV energies of fully convective field age M stars by 30.6$\pm$10.0. We calculated energy correction factors that can be used to bring the UV predictions of flare models closer in line with observations. We calculated pseudo-continuum flare temperatures that describe both the white-light and GALEX NUV emission. We measured a temperature of 10,700 K for flares from fully convective M stars after accounting for the contribution from UV line emission. We also applied our correction factors to the results of previous studies of the role of flares in abiogenesis. Our results show that M stars do not need to be as active as previously thought in order to provide the NUV flux required for prebiotic chemistry, however we note that flares will also provide more FUV flux than previously modelled.

astro-ph.SR↗

arXiv2022

Treatment Heterogeneity for Survival Outcomes

Estimation of conditional average treatment effects (CATEs) plays an essential role in modern medicine by informing treatment decision-making at a patient level. Several metalearners have been proposed recently to estimate CATEs in an effective and flexible way by re-purposing predictive machine learning models for causal estimation. In this chapter, we summarize the literature on metalearners and provide concrete guidance for their application for treatment heterogeneity estimation from randomized controlled trials' data with survival outcomes. The guidance we provide is supported by a comprehensive simulation study in which we vary the complexity of the underlying baseline risk and CATE functions, the magnitude of the heterogeneity in the treatment effect, the censoring mechanism, and the balance in treatment assignment. To demonstrate the applicability of our findings, we reanalyze the data from the Systolic Blood Pressure Intervention Trial (SPRINT) and the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study. While recent literature reports the existence of heterogeneous effects of intensive blood pressure treatment with multiple treatment effect modifiers, our results suggest that many of these modifiers may be spurious discoveries. This chapter is accompanied by survlearners, an R package that provides well-documented implementations of the CATE estimation strategies described in this work, to allow easy use of our recommendations as well as the reproduction of our numerical study.

stat.AP↗

arXiv2021

Evaluating Treatment Prioritization Rules via Rank-Weighted Average Treatment Effects

There are a number of available methods for selecting whom to prioritize for treatment, including ones based on treatment effect estimation, risk scoring, and hand-crafted rules. We propose rank-weighted average treatment effect (RATE) metrics as a simple and general family of metrics for comparing and testing the quality of treatment prioritization rules. RATE metrics are agnostic as to how the prioritization rules were derived, and only assess how well they identify individuals that benefit the most from treatment. We define a family of RATE estimators and prove a central limit theorem that enables asymptotically exact inference in a wide variety of randomized and observational study settings. RATE metrics subsume a number of existing metrics, including the Qini coefficient, and our analysis directly yields inference methods for these metrics. We showcase RATE in the context of a number of applications, including optimal targeting of aspirin to stroke patients.

stat.ME↗

arXiv2014

Development of Fiber Fabry-Perot Interferometers as Stable Near-infrared Calibration Sources for High Resolution Spectrographs

We discuss the ongoing development of single-mode fiber Fabry-Perot (FFP) Interferometers as precise astro-photonic calibration sources for high precision radial velocity (RV) spectrographs. FFPs are simple, inexpensive, monolithic units that can yield a stable and repeatable output spectrum. An FFP is a unique alternative to a traditional etalon, as the interferometric cavity is made of single-mode fiber rather than an air-gap spacer. This design allows for excellent collimation, high spectral finesse, rigid mechanical stability, insensitivity to vibrations, and no need for vacuum operation. The device we have tested is a commercially available product from Micron Optics. Our development path is targeted towards a calibration source for the Habitable-Zone Planet Finder (HPF), a near-infrared spectrograph designed to detect terrestrial-mass planets around low-mass stars, but this reference could also be used in many existing and planned fiber-fed spectrographs as we illustrate using the Apache Point Observatory Galactic Evolution Experiment (APOGEE) instrument. With precise temperature control of the fiber etalon, we achieve a thermal stability of 100 $\mu$K and associated velocity uncertainty of 22 cm s$^{-1}$. We achieve a precision of $\approx$2 m s$^{-1}$ in a single APOGEE fiber over 12 hours using this new photonic reference after removal of systematic correlations. This high precision (close to the expected photon-limited floor) is a testament to both the excellent intrinsic wavelength stability of the fiber interferometer and the stability of the APOGEE instrument design. Overall instrument velocity precision is 80 cm s$^{-1}$ over 12 hours when averaged over all 300 APOGEE fibers and after removal of known trends and pressure correlations, implying the fiber etalon is intrinsically stable to significantly higher precision.

astro-ph.IM↗