arXiv subjects

Matthew Graham

Publications and source records attributed to Matthew Graham.

At least 19 recordsLinked to original sources

arXiv2026

PRISM-AMC: Replay-Free Continual Learning for Automatic Modulation Classification

Automatic modulation classification (AMC) is a key component of cognitive wireless systems, enabling spectrum awareness and adaptive communication. Most deep learning approaches to AMC assume a fixed set of modulation classes and access to previously observed training data. In practice, modulation classes and channel conditions may evolve over time, while storage or data-retention constraints may preclude replay of raw baseband I/Q samples. We propose PRISM-AMC, a replay-free continual learning framework that mitigates catastrophic forgetting by preserving task-relevant subspaces of intermediate representations via gradient projection memory. We further analyze projected stochastic-gradient updates under an SNR-dependent gradient-noise model. PRISM-AMC is evaluated in both class-incremental and domain-incremental settings. Experiments on the MIGOU and DeepSig 2018.01A datasets show substantially reduced forgetting relative to sequential fine-tuning and elastic weight consolidation, while remaining competitive with or outperforming replay-based RSD-AMC in class-incremental settings, without storing raw I/Q samples.

cs.LG↗

arXiv2026

Discovery of a Nearby Ultra-fast Magnetic White Dwarf Propeller: Common Pathways to Magnetic Cataclysmic Variables

The fastest spinning white dwarfs (WDs), which approach break-up speeds, are found in close binary systems. In the two known magnetic WD ``propellers'', AE~Aqr and LAMOST J0240+1952, the magnetic WD spins so rapidly that it ejects the material transferred by its Roche-lobe-filling companion rather than accreting it. We report the discovery of the third such system, 1eRASS J042633.5$-$250226 (1eRASS J0426/Valeria), at only 265 pc, identified by combining SRG/eROSITA and \textit{Gaia} data via the ``X-ray Main Sequence''. TESS photometry shows AE~Aqr-like flaring and orbital modulation with $P_{\mathrm{orb}} \approx 6.62$ hr, and ZTF shows month-long high/low state changes. Phase-resolved Palomar/DBSP spectroscopy confirms the orbital period and yields a donor radial velocity of $K_2 = 202\pm7$ km s$^{-1}$. High-speed Palomar/CHIMERA photometry detects optical pulsations on three nights with the strongest detection at $P_\textrm{spin} = 26.8$ sec, though further observations are needed to confirm this as the WD spin. Spectral fitting of the donor in the low state reveals a late K-type star with $T_{\mathrm{donor}} = 4100\pm150$ K. Zeeman splitting in the optical spectrum yields a candidate field strength of $B_{\mathrm{WD}} = 26 \pm 3$ MG, meaning the system will evolve into a polar. Like other WD propellers, the system is X-ray faint ($L_X \approx 7\times 10^{30}$ erg s$^{-1}$), but is not detected in the radio, and a spin period is not detected in X-rays by SRG/eROSITA. The proximity of this system and evolutionary model constraints on the timeframe of the propeller phase yield a birth rate density of $\sim8\times10^{-16} \textrm{pc}^{-3} \textrm{yr}^{-1}$, which agrees with that of magnetic CVs, suggesting that a substantial fraction of magnetic CVs pass through a propeller phase.

astro-ph.HE↗

arXiv2026

Large-scale bioacoustic detection using semantic segmentation: a deep learning framework applied to fin whale calls in ocean-bottom seismometer recordings

Ocean-bottom seismometers (OBS), originally deployed for geophysical research, continuously record low-frequency sound for months to years across broad areas of ocean, offering a largely untapped resource for passive acoustic monitoring (PAM) of baleen whales. Realising this potential requires automated detection methods that operate reliably across the varied conditions in large sensor networks. We present a deep learning semantic segmentation framework that detects the 20-Hz notes of fin whales (Balaenoptera physalus) in OBS spectrograms, assigning each pixel a probability of belonging to a call and converting the resulting probability maps into time-frequency bounding boxes describing individual detections. We trained the model on hydrophone data from one OBS deployment in the Azores-Madeira-Canaries region and applied it without retraining to vertical-component seismometer data from a second, geographically distinct deployment, showing that a single trained model generalises across sensor types and recording environments. Applied to 378,912 h of recordings from 46 OBS sites, the detector identified 6.3 million calls, forming the largest fin whale call catalogue assembled to date, with high precision (~97%) across both deployments. The resulting catalogue resolves call timing and spectral structure accurately enough to support ecological analyses, revealing coherent seasonal shifts in three persistent inter-note interval (INI) groups across the singing season and basin-scale patterns in calling activity. By transforming existing geophysical infrastructure into a scalable sensing network, our approach substantially expands the spatial and temporal reach of PAM without new hardware investment, offering a transferable framework for tracking other low-frequency vocalising species and informing conservation planning, marine spatial management, and abundance estimation across large scales.

physics.geo-ph↗

arXiv2026

Probing Millisecond Optical Timescales with Continuous-Readout Mode Observations from ZTF

The subminute-timescale phase space is an underexplored regime of the optical sky. Sub-second optical observations are difficult to accomplish with traditional CCD exposure-readout cycles, and alternative technologies like EMCCDs and CMOS have historically limited surveys to targeted observation. Continuous-readout mode enables millisecond-timescale observations on traditional cameras by leaving the shutter open during readout, enabling high-cadence observations on all-sky surveys. We present the first results from continuous-readout mode data from the Zwicky Transient Facility. We developed a deep-learning-based data analysis pipeline and report results on 141 square degrees, or 1128 star-hours, of data, sampled at 300x/second. Our Convolutional Neural Network can retrieve optical transients of duration 9.7-29.2 milliseconds with a precision of 66% and recall of 78% for brightness changes of 10% and greater, and our subsequent light curve-based analysis has an efficiency of >94.6%. We retrieve a total of ~300,000 candidates on sources of magnitude G_{RP} < 12, of which 100 are above SNR=1.55 and were visually inspected. After accounting for noise and filtering contaminants both known and unique to this mode of observation, these 100 candidates were eliminated as being astrophysical in nature, leading to upper limits on the observed rate of ~10-ms transient events associated with point sources as well as specifically with white dwarfs and quasars. This first survey-scale characterization of the bright millisecond optical transient and variable sky establishes the rarity of short-timescale variability in the optical regime.

astro-ph.IM↗

arXiv2026

Accelerating Chemical Kinetics for Exoplanet Atmospheres using Neural Networks

Observations increasingly reveal the coupled radiative, chemical, and dynamical processes that shape exoplanet atmospheres. Interpreting these atmospheres requires models that can capture this complexity. However, multidimensional models remain fundamentally limited by computational cost, and answering key questions requires simulating the governing physical mechanisms at speeds classical methods cannot achieve. As a result, models often rely on simplifying approximations, such as equilibrium chemistry, even when those assumptions miss important effects. There is a pressing need for fast and accurate chemical kinetics solvers to model planetary atmospheres. Here we present a machine learning local-box chemical kinetics solver for exoplanet atmospheres using a residual flow-map architecture. We demonstrate that this surrogate model is several orders of magnitude faster than a classical solver, achieving microsecond-scale inference while retaining percent-level accuracy. The surrogate model covers a parameter space that spans $T=300$-$3000$ K, $P=10^{-6}$-$10^{4}$ bar, $Δt=10^{-3}$-$10^{8}$ s, and compositions ranging from $10^{-2}$ to $10^{3}$ times solar in both C/O ratio and metallicity. Our model outperforms several commonly used machine learning architectures and performs robustly under the extreme stiffness characteristic of atmospheric chemistry. The machine learning framework presented here is a flexible and efficient approach to emulating state-to-state flow-map problems that commonly arise in numerical simulations.

astro-ph.EP↗

arXiv2026

EP250827b/SN 2025wkm: An X-ray Flash-Supernova Powered by a Central Engine and Circumstellar Interaction

We present the discovery of EP250827b/SN 2025wkm, an X-ray Flash (XRF) discovered by the Einstein Probe (EP), accompanied by a broad-line Type Ic supernova (SN Ic-BL) at $z = 0.1194$. EP250827b possesses a prompt X-ray luminosity of $\sim 10^{45} \, \rm{erg \, s^{-1}}$, lasts over 1000 seconds, and has a peak energy $E_{\rm{p}} < 1.5$ keV at 90\% confidence. SN 2025wkm possesses a double-peaked optical light curve (LC), though its bolometric luminosity plateaus after its initial peak for $\sim 20$ days, consistent with a central engine injecting additional energy into the explosion. Its spectrum transitions from a blue to red continuum with clear blueshifted broad absorption features consistent with a SN Ic-BL classification. We do not detect any transient radio emission and rule out the existence of an on-axis, energetic jet $\gtrsim 10^{50}~$erg assuming a typical LGRB circumburst constant density ($n \approx 10^{-3}$--$10^{-1}~{\rm cm}^{-3}$) and microphysical parameters ($ε_{\rm e} = 0.1$ and $ε_{\rm B} = 0.01$). In the model we invoke, the collapse gives rise to a long-lived magnetar, potentially surrounded by an accretion disk. Magnetically--driven winds from the magnetar and the disk mix together and break out with a velocity $\sim 0.35c$ and interact with an extended circumstellar medium with radius $\sim 10^{13}$ cm, generating X-ray breakout emission through non-thermal free-free processes. The disk outflows and magnetar winds power blackbody photospheric emission as they cool adiabatically and thermalize, producing the first SN peak. The spin-down luminosity of the magnetar and radioactive decay of $^{56}$Ni powers the late-time emission. We end by discussing the landscape of XRF-SNe within the context of EP's recent discoveries.

astro-ph.HE↗

arXiv2026

SN~2018erx: A fast-evolving, dust-reddened Type Icn supernova with broad C II emission lines

We present the discovery and characterization of SN~2018erx (ZTF18abkmbpy), a fast-evolving, unusually red, interacting stripped-envelope supernova. Spectroscopically, SN~2018erx shows broad \ion{C}{2} emission with characteristic widths of $\sim\!3800$~km~s$^{-1}$, consistent with interaction with carbon-rich circumstellar material and a Type~Icn core-collapse SN classification. Photometrically, it evolves rapidly, rising from half-maximum to peak in 2.1~d and declining back in 3.1~d. Semi-analytical CSM-interaction modeling favors a compact, shell-like CSM with $M_{\rm CSM}\approx0.3\,M_\odot$, $R_0\approx0.7$~AU, and a low ejecta mass of $M_{\rm ej}\approx0.11\,M_\odot$. The radioactive yield is also small, with $M_{\rm Ni}\lesssim(3$--$5)\times10^{-3}\,M_\odot$, placing SN~2018erx at the low end of the H-poor distribution. At +29~d after peak, we detect a near-infrared excess consistent with pre-existing local circumstellar dust, with $M_{\rm d}\sim10^{-6}$--$10^{-5}\,M_\odot$. Together, the rapid evolution, strong local reddening, carbon-rich emission, and dust point to a multi-component circumstellar environment: a dense inner interaction region from enhanced pre-SN mass loss and an outer dusty layer from an earlier mass-loss episode roughly $10$--$200$~yr before core collapse. These properties favor an ultra-stripped core-collapse explosion of a low-mass He star in a binary system, with fallback-modified Wolf--Rayet collapse or merger-driven mass loss remaining possible alternatives. SN~2018erx provides rare insight into the mass-loss history of stripped-envelope SNe and suggests that dust-enshrouded explosions of this kind may be underrepresented in optical surveys.

astro-ph.HE↗

arXiv2026

Low-Luminosity Type IIP Supernovae from the Zwicky Transient Facility Census of the Local Universe. II: Lightcurve Analysis

The Zwicky Transient Facility Census of the Local Universe survey yielded a sample of 330 Type IIP supernovae (SNe) with well-constrained peak luminosities. In paper I (arXiv:2502.19493), we measured their luminosity function and volumetric rate. Here (paper II), we present the largest systematic study of lightcurve properties for Type IIP SNe from a volume-limited survey, analyzing a selected subset of 129 events, including 16 low-luminosity Type IIP (LLIIP) SNe with M${r,peak} \geq -16$ mag. We find that plateau slope correlates with peak brightness, with many LLIIP SNe showing positive slopes--suggesting smaller progenitor radii and distinct density profiles compared to brighter Type IIP SNe. The plateau duration shows only a weak dependence on peak brightness, likely suggesting binary interaction. One SN exhibits a plateau-to-tail drop of >3.5 mag, consistent with an electron-capture or failed SN with very low or zero nickel mass. We derive explosion and progenitor parameters of the entire Type IIP SN sample using semi-analytical and radiation-hydrodynamical models. Based on radiation-hydrodynamical model fitting, LLIIP SNe are characterized by low nickel masses (0.001-0.025 $\mathrm{M_\odot}$), low explosion energies (0.1-0.28 $\times 10^{51}$ erg), low ejecta masses ($8.1^{+0.8}_{-1.7}$ $\mathrm{M\odot}$), and ZAMS masses below 11 $\mathrm{M_\odot}$. In comparison, the full Type IIP SN sample spans a wider range with nickel masses (0.001-0.222 $\mathrm{M_\odot}$), explosion energies (0.10-4.43 $\times 10^{51}$ erg), ejecta masses (5.4-24.8 $\mathrm{M_\odot}$), and ZAMS masses (9.3-16.7 $\mathrm{M_\odot}$). We find strong correlations between peak brightness, explosion energy, and nickel mass that extend to the low-luminosity end. We conclude that LLIIP SNe represent the faint, low-energy end of the Type IIP population and originate from the lowest-mass core-collapse progenitors.

astro-ph.HE↗

arXiv2026

Electromagnetic Follow-up of the Sub-Solar Mass Gravitational Wave Candidate S251112cm: Kilonova Constraints and a Coincident IIb Supernova

On November 12th, 2025 the LIGO--Virgo--KAGRA (LVK) collaboration reported gravitational waves (GWs) from a compact object merger candidate (S251112cm) with at least one sub-solar mass component. Using the Dark Energy Camera (DECam), the Fraunhofer Telescope at Wendelstein Observatory (FTW), and the Zwicky Transient Facility (ZTF), we surveyed $56\%$ of the GW localization region beginning $2.4$~hours after the GW alert. We find no kilonova (KN) counterpart, and use radiative-transfer models to rule out $42\%$ (ZTF), $68\%$ (DECam), and $92\%$ (FTW) of the KN models as possible emission from this GW candidate. Within the recently proposed disk-fragmentation (``superkilonova'') model for generating sub-solar mass neutron star mergers from stellar core-collapse, the delay between the supernova explosion time and the GW merger time is estimated to be less than a few days. Searching this time window prior to the GW event, we identify and spectroscopically classify a IIb supernova (SN~2025adtq), with a spatial association odds ratio of $\log_{10}\mathcal{I} \approx 4.8$, a chance coincidence probability of ${\sim}2$--$9\%$, and an estimated explosion time ${\sim}2$ days prior to S251112cm. SN~2025adtq is the second Type~IIb supernova found in spatial and temporal coincidence with a sub-solar mass GW candidate, following the previously reported S250818k/SN~2025ulz association; jointly, we measure an odds ratio that favors the association hypothesis over the null, however, when conditioned on finding a coincident supernova by chance, the odds ratio disfavors association. Together, these results provide suggestive but inconclusive evidence for the superkilonova formation channel.

astro-ph.HE↗

arXiv2026

BOOM and Babamul: a real-time, multi-survey, optical alert broker system operating at scale

With the arrival of ever higher throughput wide-field surveys and a multitude of multi-messenger and multi-wavelength instruments to complement them, software capable of harnessing these associated data streams is urgently required. To meet these needs, a number of community supported alert brokers have been built, currently focused on processing of Zwicky Transient Facility (ZTF; $\sim 10^5$-$10^6$ alerts per night) with an eye towards Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST; $\sim 2 \times 10^7$ alerts per night). Building upon the system that successfully ran in production for ZTF's first seven years of operation, we introduce BOOM (Burst & Outburst Observations Monitor), an analysis framework focused on real-time, joint brokering of these alert streams. BOOM harnesses the performance of a Rust-based software stack relying on a non-relational MongoDB database combined with a Valkey in-memory processing queue and a Kafka cluster for message sharing. With this system, we demonstrate feature parity with the existing ZTF system with a throughput $\sim 7 \times$ higher. We describe the workflow that enables the real-time processing as well as the results with custom filters we have built to demonstrate the system's capabilities. In conclusion, we present the development roadmap for both BOOM and Babamul - the public-facing LSST alert broker built atop BOOM - as we begin the Rubin era.

astro-ph.IM↗

arXiv2026

AstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical Classification

Modern astronomical observatories generate a massive volume of multimodal data, creating a critical bottleneck for expert human review. While multimodal large language models (LLMs) have shown promise in interpreting complex visual and textual inputs, their ability to perform specialized scientific classification while providing interpretable reasoning remains understudied. We introduce AstroAlertBench, a comprehensive multimodal benchmark designed to evaluate LLM performance in astronomical event review along a three-stage logical chain: metadata grounding, scientific reasoning, and hierarchical classification over five categories. We use a pilot sample of 1,500 real-world alerts from the Zwicky Transient Facility (ZTF), a wide-field survey that scans the northern sky to detect transient astronomical events. On this dataset, we benchmark 13 frontier closed-source and open-weight LLMs that support visual input. Our results reveal that high accuracy does not always align with model ``honesty,'' defined as the ability to self-evaluate its reasoning, which affects its reliability as a real-world assistant. We further initialize a human-in-the-loop evaluation protocol as a precursor to future community-scale participation. Together, AstroAlertBench provides a framework for developing calibrated and interpretable astronomical assistants.

astro-ph.IM↗

arXiv2026

Semi-Supervised Learning for Lensed Quasar Detection

Lensed quasars are key to many areas of study in astronomy, offering a unique probe into the intermediate and far universe. However, finding lensed quasars has proved difficult despite significant efforts from large collaborations. These challenges have limited catalogues of confirmed lensed quasars to the hundreds, despite theoretical predictions that they should be many times more numerous. We train machine learning classifiers to discover lensed quasar candidates. By using semi-supervised learning techniques we leverage the large number of potential candidates as unlabelled training data alongside the small number of known objects, greatly improving model performance. We present our two most successful models: (1) a variational autoencoder trained on millions of quasars to reduce the dimensionality of images for input to a dense neural network classifier that can make accurate predictions and (2) a convolutional neural network trained on a mix of labelled and unlabelled data via virtual adversarial training. These models are both capable of producing high-quality candidates, as evidenced by our discovery of GRALJ140833.73+042229.98. The success of our classifier, which uses only multi-band images, is particularly exciting as it can be combined with existing classifiers, which use other data than images, to improve the classifications of both models and discover more lensed quasars.

astro-ph.IM↗

arXiv2025

SN 2021ukt: A Transitional Supernova with a Short Plateau and Persistent Interaction

We present spectroscopic and photometric observations of supernova (SN) 2021ukt, a peculiar short-plateau object that was originally identified as a Type IIn SN and later underwent an unprecedented transition to a Type Ib (possibly Type IIb) SN. The early-time light curves of SN 2021ukt exhibit a ~25 day plateau. Such a short phase of hydrogen recombination suggests a rather thin H-rich outer envelope of the progenitor star. The relatively narrow Balmer emission lines in spectra of SN 2021ukt during the first week indicate the interaction between the expanding ejecta and the immediate circumstellar material (CSM). This Hα line is observed throughout its helium-rich ejecta-dominated phase and nebular phase, suggesting persistent interaction with a radially extended CSM profile. We explore the synthetic light-curve model among grids of parameters generated by MESA+STELLA. We also compare the spectrophotometric evolution of SN 2021ukt with several well-sampled supernovae that exhibit a short plateau and persistent ejecta-CSM interaction. An estimate of the progenitor mass of SN 2021ukt is made based on the flux ratio between [Ca II] λλ 7291, 7324 and [O I] λλ 6300, 6364 during its nebular phase. Our analysis suggests that the progenitor star of SN 2021ukt has a zero-age main-sequence (ZAMS) mass of about 12 solar masses, a mass of radioactive nickel-56 synthesized in the SN ejecta of about 0.04 solar masses, and a mass of the H-rich envelope of about 0.5 solar masses. This study adds to the growing sample of transitional supernovae, reinforcing evidence for a continuum of underrepresented progenitors whose evolutionary pathways lie between those of standard SN models.

astro-ph.SR↗

arXiv2025

ZTF25abjmnps (AT2025ulz) and S250818k: A Candidate Superkilonova from a Sub-threshold Sub-Solar Gravitational Wave Trigger

On August 18, 2025, the LIGO-Virgo-KAGRA collaboration reported gravitational waves from a sub-threshold binary neutron star merger. If astrophysical, this event would have a surprisingly low chirp mass, suggesting that at least one neutron star was below a solar mass. The Zwicky Transient Facility mapped the coarse localization and discovered a transient, ZTF25abjmnps (AT2025ulz), that was spatially and temporally coincident with the gravitational wave trigger. The first week of follow-up suggested properties reminiscent of a GW170817-like kilonova. Subsequent follow-up suggests properties most similar to a young, stripped-envelope, Type IIb supernova. Although we cannot statistically rule out chance coincidence, we undertake due diligence analysis to explore the possible association between ZTF25abjmnps and S250818k. Theoretical models have been proposed wherein sub-solar neutron star(s) may form (and subsequently merge) via accretion disk fragmentation or core fission inside a core-collapse supernova i.e. a ``superkilonova". Here, we qualitatively discuss our multi-wavelength dataset in the context of the superkilonova picture. Future higher significance gravitational wave detections of sub-solar neutron star mergers with extensive electromagnetic follow-up would conclusively resolve this tantalizing multi-messenger association.

astro-ph.HE↗

arXiv2025

Accelerating Radiative Transfer for Planetary Atmospheres by Orders of Magnitude with a Transformer-Based Machine Learning Model

Radiative transfer calculations are essential for modeling planetary atmospheres. However, standard methods are computationally demanding and impose accuracy-speed trade-offs. High computational costs force numerical simplifications in large models (e.g., General Circulation Models) that degrade the accuracy of the simulation. Radiative transfer calculations are an ideal candidate for machine learning emulation: fundamentally, it is a well-defined physical mapping from a static atmospheric profile to the resulting fluxes, and high-fidelity training data can be created from first principles calculations. We developed a radiative transfer emulator using an encoder-only transformer neural network architecture, trained on 1D profiles representative of solar-composition hot Jupiter atmospheres. Our emulator reproduced bolometric two-stream layer fluxes with mean test set errors of ~1% compared to the traditional method and achieved speedups of 100x. Emulating radiative transfer with machine learning opens up the possibility for faster and more accurate routines within planetary atmospheric models such as GCMs.

astro-ph.EP↗

arXiv2025

LIGO/Virgo/KAGRA neutron star merger candidate S250206dm: Zwicky Transient Facility observations

We present the searches conducted with the Zwicky Transient Facility (ZTF) in response to S250206dm, a bona fide event with a false alarm rate of one in 25 years, detected by the International Gravitational Wave Network (IGWN). Although the event is significant, the nature of the compact objects involved remains unclear, with at least one likely neutron star. ZTF covered 68% of the localization region, though we did not identify any likely optical counterpart. We describe the ZTF strategy, potential candidates, and the observations that helped rule out candidates, including sources circulated by other collaborations. Similar to Ahumada et al. 2024, we perform a frequentist analysis, using simsurvey, as well as Bayesian analysis, using nimbus, to quantify the efficiency of our searches. We find that, given the nominal distance to this event of 373$\pm$104 Mpc, our efficiencies are above 10% for KNe brighter than $-17.5$ absolute magnitude. Assuming the optical counterpart known as kilonova (KN) lies within the ZTF footprint, our limits constrain the brightest end of the KN parameter space. Through dedicated radiative transfer simulations of KNe from binary neutron star (BNS) and black hole-neutron star (BHNS) mergers, we exclude parts of the BNS KN parameter space. Up to 35% of the models with high wind ejecta mass ($M_{\rm wind} \approx 0.13$ M$_{\odot}$) are ruled out when viewed face-on ($\cosθ_{\rm obs} = 1.0$). Finally, we present a joint analysis using the combined coverage from ZTF and the Gravitational Wave Multimessenger Dark Energy Camera Survey (GW-MMADS). The joint observations cover 73% of the localization region, and the combined efficiency has a stronger impact on rising and slowly fading models, allowing us to rule out 55% of the high-mass KN models viewed face-on.

astro-ph.HE↗

arXiv2025

Infrared-Selected Active Galactic Nuclei in the Kepler Fields

We utilized the Edelson and Malkan (2012) and Stern et al. (2012) selection techniques and other methods to identify AGN candidates that were monitored during the Kepler prime and K2 missions. Subsequent to those observations, we obtained 125 long-slit optical spectra with the Lick 3-m telescope, 58 spectra with the Palomar 5-m telescope, and three with the Keck 10-m telescope to test these identifications. Of these 186 AGN candidates, 105 were confirmed as Type 1 AGN and 35 as Type 2 AGN, while the remaining 46 were found to have other identifications (e.g., stars and normal galaxies). This indicated an overall reliability of about 75%, while the two main methods had much higher reliability, 87%-96%. The spectra indicated redshifts out to z = 3.4. Then, we examined the AGN sample properties through the Baldwin, Phillips & Terlevich diagram and compared the AGN's spectral energy distributions (SEDs) with those from the literature. We found that our sample yielded the same AGN population as those identified through other methods, such as optical spectroscopy.

astro-ph.GA↗

arXiv2025

Building Machine Learning Challenges for Anomaly Detection in Science

Scientific discoveries are often made by finding a pattern or object that was not predicted by the known rules of science. Oftentimes, these anomalous events or objects that do not conform to the norms are an indication that the rules of science governing the data are incomplete, and something new needs to be present to explain these unexpected outliers. The challenge of finding anomalies can be confounding since it requires codifying a complete knowledge of the known scientific behaviors and then projecting these known behaviors on the data to look for deviations. When utilizing machine learning, this presents a particular challenge since we require that the model not only understands scientific data perfectly but also recognizes when the data is inconsistent and out of the scope of its trained behavior. In this paper, we present three datasets aimed at developing machine learning-based anomaly detection for disparate scientific domains covering astrophysics, genomics, and polar science. We present the different datasets along with a scheme to make machine learning challenges around the three datasets findable, accessible, interoperable, and reusable (FAIR). Furthermore, we present an approach that generalizes to future machine learning challenges, enabling the possibility of large, more compute-intensive challenges that can ultimately lead to scientific discovery.

cs.LG↗