Search NASA⌕ Search

SEARCH · Search NASA

Results for “PREDICTION”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 829 records · Page 46

Budget constrained machine learning for early prediction of adverse outcomes for COVID-19 patients

Abstract The combination of machine learning (ML) and electronic health records (EHR) data may be able to improve outcomes of hospitalized COVID-19 patients through improved risk stratification and patient outcome prediction. However, in resource constrained environments the clinical utility of such data-driven predictive tools may be limited by the cost or unavailability of certain laboratory tests. We leveraged EHR data to develop an ML-based tool for predicting adverse outcomes that optimizes clinical utility under a given cost structure. We further gained insights into the decision-making process of the ML models through an explainable AI tool. This cohort study was performed using deidentified EHR data from COVID-19 patients from ProMedica Health System in northwest Ohio and southeastern Michigan. We tested the performance of various ML approaches for predicting either increasing ventilatory support or mortality. We performed post hoc analysis to obtain optimal feature sets under various budget constraints. We demonstrate that it is possible to achieve a significant reduction in cost at the expense of a small reduction in predictive performance. For example, when predicting ventilation, it is possible to achieve a 43% reduction in cost with only a 3% reduction in performance. Similarly, when predicting mortality, it is possible to achieve a 50% reduction in cost with only a 1% reduction in performance. This study presents a quick, accurate, and cost-effective method to evaluate risk of deterioration for patients with SARS-CoV-2 infection at the time of clinical evaluation.

97 MATHEMATICS AND COMPUTING↗

The predictive power of phylogeny on growth rates in soil bacterial communities

Abstract Predicting ecosystem function is critical to assess and mitigate the impacts of climate change. Quantitative predictions of microbially mediated ecosystem processes are typically uninformed by microbial biodiversity. Yet new tools allow the measurement of taxon-specific traits within natural microbial communities. There is mounting evidence of a phylogenetic signal in these traits, which may support prediction and microbiome management frameworks. We investigated phylogeny-based trait prediction using bacterial growth rates from soil communities in Arctic, boreal, temperate, and tropical ecosystems. Here we show that phylogeny predicts growth rates of soil bacteria, explaining an average of 31%, and up to 58%, of the variation within ecosystems. Despite limited overlap in community composition across these ecosystems, shared nodes in the phylogeny enabled ancestral trait reconstruction and cross-ecosystem predictions. Phylogenetic relationships could explain up to 38% (averaging 14%) of the variation in growth rates across the highly disparate ecosystems studied. Our results suggest that shared evolutionary history contributes to similarity in the relative growth rates of related bacteria in the wild, allowing phylogeny-based predictions to explain a substantial amount of the variation in taxon-specific functional traits, within and across ecosystems.

Walkup, Jeth (ORCID:0000000279082963)↗

ZeoNet: 3D convolutional neural networks for predicting adsorption in nanoporous zeolites

Zeolites are one of the most widely used materials in the chemical industry due to their nanometer-sized pores that can adsorb and react upon molecules selectively. With hundreds of known framework topologies and hundreds of thousands of computationally predicted structures, the ability to rapidly predict zeolite performance allows researchers to prioritize their efforts on the most promising structures for a given application. Although the accuracy of forcefield-based atomistic simulations has advanced significantly in the past two decades, these simulations can be computationally expensive, especially for long-chain, complex molecules. Here, we present ZeoNet, a representation learning framework using convolutional neural networks (ConvNets) and 3D volumetric representations for predicting adsorption in zeolites. ZeoNet was trained on the task of predicting Henry's constants for adsorption, k H , of n-octadecane in more than 330 000 known and predicted zeolite materials. Employing a 3D grid based on the distances to solvent-accessible surfaces, a volumetric representation that can be generated efficiently, the best-performing ZeoNet achieved a correlation coefficient r 2 = 0.977 and a mean-squared error MSE = 3.8 in ln k H , which corresponds to an error of 9.3 kJ mol -1 in adsorption free energy. In comparison, a model based on hand-designed geometric features has values of r 2 = 0.783 and MSE = 35.7. ZeoNet is also relatively efficient and can process ≈8 structures per second on an Nvidia RTX 2080TI GPU, orders of magnitude faster than forcefield-based simulations. A systematic analysis was conducted to investigate how the choice of ConvNet architectures, the linear dimension (L) and spatial resolution (Δd) of the distance grids, batch size, optimizer, and learning rate impact the model performance. We found that ConvNets based on the ResNet architecture offer the best tradeoff between expressiveness and efficiency. The performance for all models reaches a plateau at L = 30–45 Å and depends less sensitively on grid resolution, with a small benefit around Δd = 0.30–0.45 Å. Finally, saliency maps were visualized to identify which regions of the materials contributed the most to model predictions. It was found, interestingly, that the predictions are driven primarily by the accessible pore volume rather than the region occupied by the framework atoms.

36 MATERIALS SCIENCE↗

Predicting the effect of angular momentum on the dissociation dynamics of highly rotationally excited radical intermediates

We present a model which accurately predicts the net speed distributions of products resulting from the unimolecular decomposition of rotationally excited radicals. The radicals are produced photolytically from a halogenated precursor under collision-free conditions so they are not in a thermal distribution of rotational states. The accuracy relies on the radical dissociating with negligible energetic barrier beyond the endoergicity. We test the model predictions using previous velocity map imaging and crossed laser-molecular beam scattering experiments that photolytically generated rotationally excited CD{sub 2}CD{sub 2}OH and C{sub 3}H{sub 6}OH radicals from brominated precursors; some of those radicals then undergo further dissociation to CD{sub 2}CD{sub 2} + OH and C{sub 3}H{sub 6} + OH, respectively. We model the rotational trajectories of these radicals, with high vibrational and rotational energy, first near their equilibrium geometry, and then by projecting each point during the rotation to the transition state (continuing the rotational dynamics at that geometry). This allows us to accurately predict the recoil velocity imparted in the subsequent dissociation of the radical by calculating the tangential velocities of the CD{sub 2}CD{sub 2}/C{sub 3}H{sub 6} and OH fragments at the transition state. The model also gives a prediction for the distribution of angles between the dissociation fragments’ velocity vectors and the initial radical’s velocity vector. These results are used to generate fits to the previously measured time-of-flight distributions of the dissociation fragments; the fits are excellent. The results demonstrate the importance of considering the precession of the angular velocity vector for a rotating radical. We also show that if the initial angular momentum of the rotating radical lies nearly parallel to a principal axis, the very narrow range of tangential velocities predicted by this model must be convoluted with a J = 0 recoil velocity distribution to achieve a good result. The model relies on measuring the kinetic energy release when the halogenated precursor is photodissociated via a repulsive excited state but does not include any adjustable parameters. Even when different conformers of the photolytic precursor are populated, weighting the prediction by a thermal conformer population gives an accurate prediction for the relative velocity vectors of the fragments from the highly rotationally excited radical intermediates.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Earthquake prediction with electromagnetic phenomena

Short-term earthquake (EQ) prediction is defined as prospective prediction with the time scale of about one week, which is considered to be one of the most important and urgent topics for the human beings. If this short-term prediction is realized, casualty will be drastically reduced. Unlike the conventional seismic measurement, we proposed the use of electromagnetic phenomena as precursors to EQs in the prediction, and an extensive amount of progress has been achieved in the field of seismo-electromagnetics during the last two decades. This paper deals with the review on this short-term EQ prediction, including the impossibility myth of EQs prediction by seismometers, the reason why we are interested in electromagnetics, the history of seismo-electromagnetics, the ionospheric perturbation as the most promising candidate of EQ prediction, then the future of EQ predictology from two standpoints of a practical science and a pure science, and finally a brief summary.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Quasi-linear gyrokinetic predictions of the Coriolis momentum pinch in National Spherical Torus Experiment

This paper presents quasi-linear gyrokinetic predictions of the Coriolis momentum pinch for low aspect-ratio National Spherical Torus Experiment (NSTX) H-modes where previous experimental measurements were focused. Local, linear calculations predict that in the region of interest (just outside the mid-radius) of these relatively high-beta plasmas, profiles are most unstable to microtearing modes that are only effective in transporting electron energy. However, sub-dominant electromagnetic and electrostaticballooning modes are also unstable, which are effective at transporting energy, particles, and momentum. The quasi-linear prediction of transport from these weaker ballooning modes, assuming they contribute transport in addition to that from microtearing modes in a nonlinear turbulent state, leads to a very small or outward convection of momentum, inconsistent with the experimentally measured inward pinch, and opposite to predictions in conventional aspect ratio tokamaks. Additional predictions of a low beta L-mode plasma, unstable to more traditional electrostatic ion temperature gradient-trapped electron mode instability, show that the Coriolis pinch is inward but remains relatively weak and insensitive to many parameter variations. The weak or outward pinch predicted in NSTX plasmas appears to be at least partially correlated to changes in the parallel mode structure that occur at a finite beta and low aspect ratio, as discussed in previous theories. The only conditions identified where a stronger inward pinch is predicted occur either in the purely electrostatic limit or if the aspect ratio is increased. Lastly, as the Coriolis pinch cannot explain the measured momentum pinch, additional theoretical momentum transport mechanisms are discussed that may be potentially important.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Predicting rotation for ITER via studies of intrinsic torque and momentum transport in DIII-D

Here, experiments at the DIII-D tokamak have used dimensionless parameter scans to investigate the dependencies of intrinsic torque and momentum transport in order to inform a prediction of the rotation profile in ITER. Measurements of intrinsic torque profiles and momentum confinement time in dimensionless parameter scans of normalized gyroradius and collisionality are used to predict the amount of intrinsic rotation in the pedestal of ITER. Additional scans of T e /T i and safety factor are used to determine the accuracy of momentum flux predictions of the quasi-linear gyrokinetic code TGLF. In these scans, applications of modulated torque are used to measure the incremental momentum diffusivity, and results are consistent with the E x B shear suppression of turbulent transport. These incremental transport measurements are also compared with the TGLF results. In order to form a prediction of the rotation profile for ITER, the pedestal prediction is used as a boundary condition to a simulation that uses TGLF to determine the transport in the core of the plasma. The predicted rotation is ≈20 krad/s in the core, lower than in many current tokamak operating scenarios. TGLF predictions show that this rotation is still significant enough to have a strong effect on confinement via E x B shear.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Predicting the thermodynamic stability of double-perovskite halides from density functional theory

Recently, a series of double-perovskite halide compounds such as Cs 2 AgBiCl 6 and Cs 2 AgBiBr 6 have attracted intensive interest as promising alternatives to the solar absorber material CH 3 NH 3 PbI 3 because they are Pb-free and may exhibit enhanced stability. The thermodynamic stability of a number of double-perovskite halides has been predicted based on density functional theory (DFT) calculations of compound formation energies. In this paper, we found that the stability prediction can be dependent on the approximations used for the exchange-correlation functionals, e.g., the DFT calculations using the widely used Perdew, Burke, Ernzerhof (PBE) functional predict that Cs 2 AgBiBr 6 is thermodynamically unstable against phase-separation into the competing phases such as AgBr, Cs 2 AgBr 3 , Cs 3 Bi 2 Br 9 , etc., obviously inconsistent with the good stability observed experimentally. The incorrect prediction by the PBE calculation results from its failure to predict the correct ground-state structures of AgBr, AgCl, and CsCl. By contrast, the DFT calculations based on local density approximation, optB86b-vdW, and optB88-vdW functionals predict the ground-state structures of these binary halides correctly. Furthermore, the optB88-vdW functional is found to give the most accurate description of the lattice constants of the double-perovskite halides and their competing phases. Given these two aspects, we suggest that the optB88-vdW functional should be used for predicting thermodynamic stability in the future high-throughput computational material design or the construction of the Materials Genome database for new double-perovskite halides. As a result, using different exchange-correlation functionals has little influence on the dispersion of the conduction and the valence bands near the electronic bandgap; however, the calculated bandgap can be affected indirectly by the optimized lattice constant, which varies for different functionals.

36 MATERIALS SCIENCE↗

Predictability as a probe of manifest and latent physics: The case of atomic scale structural, chemical, and polarization behaviors in multiferroic Sm-doped BiFeO 3

The predictability of a certain effect or phenomenon is often equated with the knowledge of relevant physical laws, typically understood as a functional or numerically derived relationship between the observations and known states of the system. Correspondingly, observations inconsistent with prior knowledge can be used to derive new knowledge on the nature of the system or indicate the presence of yet unknown mechanisms. In this work, we explore the applicability of Gaussian processes (GP) to establish predictability and uncertainty of local behaviors from multimodal observations, providing an alternative to this classical paradigm. Using atomic resolution scanning transmission electron microscopy (STEM) of multiferroic Sm-doped BiFeO 3 across a broad composition range, we directly visualize the atomic structure and structural, physical, and chemical order parameter fields for the material. GP regression is used to establish the predictability of the local polarization field from different groups of parameters, including the adjacent polarization values and several combinations of physical and chemical descriptors, including lattice parameters, column intensities, etc. We observe that certain elements of microstructure, including charged and uncharged domain walls and interfaces with the substrate, are best predicted with specific combinations of descriptors, and this predictability and associated uncertainties are consistent across the composition series. The associated generative physical mechanisms are discussed. It is also found that certain parameter combinations tend to predict the orthorhombic phase in the cases where rhombohedral phase is observed, suggesting a potential role of clamping and confinement phenomena in phase equilibrium in Sm-BiFeO 3 system close to morphotropic phase boundary. We argue that predictability and uncertainty in observational data offer a new pathway to probe the physics of condensed matter systems from multimodal local observations.

74 ATOMIC AND MOLECULAR PHYSICS↗

Materials representation and transfer learning for multi-property prediction

The adoption of machine learning in materials science has rapidly transformed materials property prediction. Hurdles limiting full capitalization of recent advancements in machine learning include the limited development of methods to learn the underlying interactions of multiple elements as well as the relationships among multiple properties to facilitate property prediction in new composition spaces. To address these issues, we introduce the Hierarchical Correlation Learning for Multi-property Prediction (H-CLMP) framework that seamlessly integrates: (i) prediction using only a material's composition, (ii) learning and exploitation of correlations among target properties in multi-target regression, and (iii) leveraging training data from tangential domains via generative transfer learning. The model is demonstrated for prediction of spectral optical absorption of complex metal oxides spanning 69 three-cation metal oxide composition spaces. H-CLMP accurately predicts non-linear composition-property relationships in composition spaces for which no training data are available, which broadens the purview of machine learning to the discovery of materials with exceptional properties. This achievement results from the principled integration of latent embedding learning, property correlation learning, generative transfer learning, and attention models. The best performance is obtained using H-CLMP with transfer learning [H-CLMP(T)] wherein a generative adversarial network is trained on computational density of states data and deployed in the target domain to augment prediction of optical absorption from composition. H-CLMP(T) aggregates multiple knowledge sources with a framework that is well suited for multi-target regression across the physical sciences.

36 MATERIALS SCIENCE↗

Accelerating defect predictions in semiconductors using graph neural networks

First-principles computations reliably predict the energetics of point defects in semiconductors but are constrained by the expense of using large supercells and advanced levels of theory. Machine learning models trained on computational data, especially ones that sufficiently encode defect coordination environments, can be used to accelerate defect predictions. Here, we develop a framework for the prediction and screening of native defects and functional impurities in a chemical space of group IV, III–V, and II–VI zinc blende semiconductors, powered by crystal Graph-based Neural Networks (GNNs) trained on high-throughput density functional theory (DFT) data. Using an innovative approach of sampling partially optimized defect configurations from DFT calculations, we generate one of the largest computational defect datasets to date, containing many types of vacancies, self-interstitials, anti-site substitutions, impurity interstitials and substitutions, as well as some defect complexes. We applied three types of established GNN techniques, namely crystal graph convolutional neural network, materials graph network, and Atomistic Line Graph Neural Network (ALIGNN), to rigorously train models for predicting defect formation energy (DFE) in multiple charge states and chemical potential conditions. We find that ALIGNN yields the best DFE predictions with root mean square errors around 0.3 eV, which represents a prediction accuracy of 98% given the range of values within the dataset, improving significantly on the state-of-the-art. We further show that GNN-based defective structure optimization can take us close to DFT-optimized geometries at a fraction of the cost of full DFT. The current models are based on the semi-local generalized gradient approximation-Perdew–Burke–Ernzerhof (PBE) functional but are highly promising because of the correlation of computed energetics and defect levels with higher levels of theory and experimental data, the accuracy and necessity of discovering novel metastable and low energy defect structures at the PBE level of theory before advanced methods could be applied, and the ability to train multi-fidelity models in the future with new data from non-local functionals. The DFT-GNN models enable prediction and screening across thousands of hypothetical defects based on both unoptimized and partially optimized defective structures, helping identify electronically active defects in technologically important semiconductors.

Rahman, Md Habibur (ORCID:000000027705984X)↗

Understanding the effect of density functional choice and van der Waals treatment on predicting the binding configuration, loading, and stability of amine-grafted metal organic frameworks

Metal organic frameworks (MOFs) are crystalline, three-dimensional structures with high surface areas and tunable porosities. Made from metal nodes connected by organic linkers, the exact properties of a given MOF are determined by node and linker choice. MOFs hold promise for numerous applications, including gas capture and storage. M2(4,4′-dioxidobiphenyl-3,3′-dicarboxylate)—henceforth simply M2(dobpdc), with M = Mg, Mn, Fe, Co, Ni, Cu, or Zn—is regarded as one of the most promising structures for CO2 capture applications. Further modification of the MOF with diamines or tetramines can significantly boost gas species selectivity, a necessity for the ultra-dilute CO2 concentrations in the direct-air capture of CO2. There are countless potential diamines and tetramines, paving the way for a vast number of potential sorbents to be probed for CO2 adsorption properties. The number of amines and their configuration in the MOF pore are key drivers of CO2 adsorption capacity and kinetics, and so a validation of computational prediction of these quantities is required to suitably use computational methods in the discovery and screening of amine-functionalized sorbents. In this work, we study the predictive accuracy of density functional theory and related calculations on amine loading and configuration for one diamine and two tetramines. In particular, we explore the Perdew–Burke–Ernzerhof (PBE) functional and its formulation for solids (PBEsol) with and without the Grimme-D2 and Grimme-D3 pairwise corrections (PBE+D2/3 and PBEsol+D2/3), two revised PBE functionals with the Grimme-D2 and Grimme-D3 pairwise corrections (RPBE+D2/3 and revPBE+D2/3), and the nonlocal van der Waals correlation (vdW-DF2) functional. We also investigate a universal graph deep learning interatomic potential’s (M3GNet) predictive accuracy for loading and configuration. These results allow us to identify a useful screening procedure for configuration prediction that has a coarse component for quick evaluation and a higher accuracy component for detailed analysis. Our general observation is that the neural network-based potential can be used as a high-level and rapid screening tool, whereas PBEsol+D3 gives a completely qualitatively predictive picture across all systems studied, and can thus be used for high accuracy motif predictions. We close by briefly exploring the predictions of relative thermal stability for the different functionals and dispersion corrections.

Chemistry↗

Predicting climate change impacts on poikilotherms using physiologically guided species abundance models

Poikilothermic animals comprise most species on Earth and are especially sensitive to changes in environmental temperatures. Species conservation in a changing climate relies upon predictions of species responses to future conditions, yet predicting species responses to climate change when temperatures exceed the bounds of observed data is fraught with challenges. We present a physiologically guided abundance (PGA) model that combines observations of species abundance and environmental conditions with laboratory-derived data on the physiological response of poikilotherms to temperature to predict species geographical distributions and abundance in response to climate change. The model incorporates uncertainty in laboratory-derived thermal response curves and provides estimates of thermal habitat suitability and extinction probability based on site-specific conditions. We show that temperature-driven changes in distributions, local extinction, and abundance of cold, cool, and warm-adapted species vary substantially when physiological information is incorporated. Notably, cold-adapted species were predicted by the PGA model to be extirpated in 61% of locations that they currently inhabit, while extirpation was never predicted by a correlative niche model. Failure to account for species-specific physiological constraints could lead to unrealistic predictions under a warming climate, including underestimates of local extirpation for cold-adapted species near the edges of their climate niche space and overoptimistic predictions of warm-adapted species.

54 ENVIRONMENTAL SCIENCES↗

The Role of Nuclear Data Sensitivities in Prompt α-Eigenvalue Predictions of Delayed Critical Benchmarks

Alpha (α) eigenvalues, which describe the logarithmic time derivative of the neutron population in a multiplying system, are integral to time-dependent behavior and diagnostic applications. However, uncertainties in the evaluated nuclear data can significantly impact the accuracy of transport simulations for such quantities. This work explores the use of machine learning models to predict two key outputs, α-eigenvalues and keff bias, using input features derived from α-eigenvalue sensitivities to nuclear data. The criticality safety benchmark models used in this study come from the International Handbook of Evaluated Criticality Safety Benchmark Experiments. Three models, random forest, XGBoost, and NGBoost, are trained on both energy-resolved and energy-summed α sensitivities. For the α-eigenvalue bias prediction, NGBoost achieved the highest R 2 (0.9476) using energy-resolved features, while XGBoost performed best using summed sensitivities. In contrast, when predicting the keff bias, all the models showed moderate predictive capability (best R 2 ≈ 0.72), as the mapping from the static α-sensitivities to the static keff bias was less direct. SHAP (SHapley Additive exPlanations) analysis was used to interpret the model predictions. Across both prediction tasks, the features associated with neutron capture [H-1 (n, γ)], uranium scattering reactions (such as 235 U elastic/inelastic), and actinide capture/fission reactions (such as 239 Pu and 234 U) were consistently identified as the most impactful. This highlights the key role of specific nuclear reactions and energy ranges in shaping both time-dependent and steady-state criticality behavior. These results demonstrated that α-sensitivities, despite being computed for time-dependent metrics, can provide valuable insights for predicting both α-eigenvalues and the keff bias. Moreover, machine learning models offer a promising pathway for uncovering important nuclear data dependencies and guiding future data evaluation efforts.

Nuclear data↗

Advances in prediction of tokamak experiments with theory-based models

The successful validation of theory-based models of transport, MHD stability, heating and current drive, with tokamak measurements over the last 20 years, has laid the foundation for a new era where these models can be routinely used in a "predict first" approach to design and predict the outcomes of experiments on tokamaks today. The capability to predict the plasma confinement and core profiles with a quantified uncertainty, based on a multi-machine, international, database of experience, will provide confidence that a proposed discharge will remain within the operational limits of the tokamak. Developing this predictive capability for the first generation of burning plasma devices, beginning with ITER, and progressing to tokamak demonstration reactors, is a critical mission of fusion energy research. Major advances have been made implementing this predict first methodology on today's tokamaks. In this work, an overview of several of these recent advances will be presented, providing the integrated modeling foundations of the experimental successes. The first steps to include boundary plasmas, and tokamak control systems, have been made. A commitment to predicting experiments as part of the planning process is needed in order to collect predictive accuracy data and evolve the models and software into a robust whole discharge pulse design simulator.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

RNAG: a new Gibbs sampler for predicting RNA secondary structure for unaligned sequences

Motivation: RNA secondary structure plays an important role in the function of many RNAs, and structural features are often key to their interaction with other cellular components. Thus, there has been considerable interest in the prediction of secondary structures for RNA families. In this article, we present a new global structural alignment algorithm, RNAG, to predict consensus secondary structures for unaligned sequences. It uses a blocked Gibbs sampling algorithm, which has a theoretical advantage in convergence time. This algorithm iteratively samples from the conditional probability distributions P(Structure | Alignment) and P(Alignment | Structure). Not surprisingly, there is considerable uncertainly in the high-dimensional space of this difficult problem, which has so far received limited attention in this field. We show how the samples drawn from this algorithm can be used to more fully characterize the posterior space and to assess the uncertainty of predictions. Results: Our analysis of three publically available datasets showed a substantial improvement in RNA structure prediction by RNAG over extant prediction methods. Additionally, our analysis of 17 RNA families showed that the RNAG sampled structures were generally compact around their ensemble centroids, and at least 11 families had at least two well-separated clusters of predicted structures. In general, the distance between a reference structure and our predicted structure was large relative to the variation among structures within an ensemble. Availability: The Perl implementation of the RNAG algorithm and the data necessary to reproduce the results described in Sections 3.1 and 3.2 are available at http://ccmbweb.ccv.brown.edu/rnag.html

59 BASIC BIOLOGICAL SCIENCES↗

Uncertainty quantification of the virial black hole mass with conformal prediction

Precise measurements of the black hole mass are essential to gain insight on the black hole and host galaxy co-evolution. A direct measure of the black hole mass is often restricted to nearest galaxies and instead, an indirect method using the single-epoch virial black hole mass estimation is used for objects at high redshifts. However, this method is subjected to biases and uncertainties as it is reliant on the scaling relation from a small sample of local active galactic nuclei. In this study, we propose the application of conformalized quantile regression (CQR) to quantify the uncertainties of the black hole predictions in a machine learning setting. We compare CQR with various prediction interval techniques and demonstrated that CQR can provide a more useful prediction interval indicator. In contrast to baseline approaches for prediction interval estimation, we show that the CQR method provides prediction intervals that adjust to the black hole mass and its related properties. That is it yields a tighter constraint on the prediction interval (hence more certain) for a larger black hole mass, and accordingly, bright and broad spectral line width source. Using a combination of neural network model and CQR framework, the recovered virial black hole mass predictions and uncertainties are comparable to those measured from the Sloan Digital Sky Survey. The code is publicly available.

79 ASTRONOMY AND ASTROPHYSICS↗

Emission line predictions for mock galaxy catalogues: a new differentiable and empirical mapping from DESI

ABSTRACT We present a simple, differentiable method for predicting emission line strengths from rest-frame optical continua using an empirically determined mapping. Extensive work has been done to develop mock galaxy catalogues that include robust predictions for galaxy photometry, but reliably predicting the strengths of emission lines has remained challenging. Our new mapping is a simple neural network implemented using the JAX Python automatic differentiation library. It is trained on Dark Energy Spectroscopic Instrument Early Release data to predict the equivalent widths (EWs) of the eight brightest optical emission lines (including H α, H β, [O ii], and [O iii]) from a galaxy’s rest-frame optical continuum. The predicted EW distributions are consistent with the observed ones when noise is accounted for, and we find Spearman’s rank correlation coefficient ρs > 0.87 between predictions and observations for most lines. Using a non-linear dimensionality reduction technique, we show that this is true for galaxies across the full range of observed spectral energy distributions. In addition, we find that adding measurement uncertainties to the predicted line strengths is essential for reproducing the distribution of observed line-ratios in the BPT diagram. Our trained network can easily be incorporated into a differentiable stellar population synthesis pipeline without hindering differentiability or scalability with GPUs. A synthetic catalogue generated with such a pipeline can be used to characterize and account for biases in the spectroscopic training sets used for training and calibration of photo-z’s, improving the modelling of systematic incompleteness for the Rubin Observatory LSST and other surveys.

79 ASTRONOMY AND ASTROPHYSICS↗