arXiv subjects

Arjun Suresh

Publications and source records attributed to Arjun Suresh.

11 recordsLinked to original sources

arXiv2026

AGN Feedback Models and AGN Demographics II: Comparing Predictions of Radiative and Total Feedback to Observations

We evaluate the radiative-mode active galactic nucleus (AGN) feedback models of EAGLE, SIMBA, and TNG100 by comparing their predictions for the AGN-host galaxy relationship to new observational constraints. Owing to incomplete knowledge of the underlying physics, these models differ substantially, and it remains unclear whether any of them accurately reflect reality. In a previous study, based on the demographics of narrow line AGN, we constrained $F_{\mathrm{AGN}}$, the intrinsic fraction of galaxies hosting a radiative AGN with Eddington ratio $\lambda > 10^{-3}$, as a function of host stellar mass ($M_*$) and specific star formation rate (sSFR). Observationally, $F_{\mathrm{AGN}}$ declines strongly with $M_*$ for quiescent galaxies, while remaining approximately constant for star-forming systems. In this study, we find that none of the simulations reproduce these trends even qualitatively, indicating a mismatch between the simulated and observed radiative AGN populations. Additionally, since EAGLE does not explicitly distinguish between radio and radiative modes, we compare its predictions for $F_{\mathrm{AGN}}(M_*, \mathrm{sSFR})$ to our novel measurements of the combined radiative and radio mode AGN fractions. For EAGLE's constant coupling efficiency $\epsilon_{\mathrm{f}} = 0.15$, and for the adopted standard bolometric and jet-power conversion factors in the observations, the total AGN feedback energy in EAGLE is broadly consistent with observations, though with Eddington ratios a factor of 10 greater than is observed. More detailed comparisons of EAGLE's assumptions with observations are therefore required before drawing firm conclusions.

astro-ph.GA↗

arXiv2025

The Eddington Ratio Distribution of Narrow Line Active Galactic Nuclei

We measure the Eddington ratio distribution of local optical narrow-line active galactic nuclei (AGN) as a function of host galaxy properties, as a potential test of galaxy formation theories of AGN feedback. We extract central emission-line fluxes using data from the Mapping Nearby Galaxies at APO (MaNGA) sample of the Sloan Digital Sky Survey IV Data Release 17. Using the line ratio diagnostic techniques of Ji & Yan (2020), we identify AGN galaxies and determine their H$\beta$ and [OIII] line luminosities. For all galaxies not identified as AGN, we determine the threshold line luminosity they would have needed to be identified as AGN. These luminosity thresholds allow us to account for selection effects that otherwise would lead to strongly biased results. From the H$\beta$ luminosities and luminosity detection thresholds, accounting for selection effects, we measure the luminosity and Eddington ratio distributions of Seyferts as a function of specific star formation rate (sSFR) and stellar mass. Defining $F_{\rm AGN}$ as the occurrence rate of AGN above a fixed Eddington ratio of $10^{-3}$, we find that $F_{\rm AGN}$ is constant or increasing with stellar mass for star forming galaxies and declines strongly with stellar mass for quiescent galaxies. At stellar masses $\log_{10} M_\ast > 10.25$, the occurrence rate increases monotonically with sSFR. At low statistical significance, in our lowest mass bins $9.25 < \log_{10} M_\ast < 10.25$, $F_{\rm AGN}$ peaks at intermediate sSFR. These patterns reveal a complicated dependence of AGN activity on galaxy properties for theoretical models to explain.

astro-ph.GA↗

arXiv2025

Safer in Translation? Presupposition Robustness in Indic Languages

Increasingly, more and more people are turning to large language models (LLMs) for healthcare advice and consultation, making it important to gauge the efficacy and accuracy of the responses of LLMs to such queries. While there are pre-existing medical benchmarks literature which seeks to accomplish this very task, these benchmarks are almost universally in English, which has led to a notable gap in existing literature pertaining to multilingual LLM evaluation. Within this work, we seek to aid in addressing this gap with Cancer-Myth-Indic, an Indic language benchmark built by translating a 500-item subset of Cancer-Myth, sampled evenly across its original categories, into five under-served but widely used languages from the subcontinent (500 per language; 2,500 translated items total). Native-speaker translators followed a style guide for preserving implicit presuppositions in translation; items feature false presuppositions relating to cancer. We evaluate several popular LLMs under this presupposition stress.

cs.CL↗

arXiv2025

EREBUS: End-to-end Robust Event Based Underwater Simulation

The underwater domain presents a vast array of challenges for roboticists and computer vision researchers alike, such as poor lighting conditions and high dynamic range scenes. In these adverse conditions, traditional vision techniques struggle to adapt and lead to suboptimal performance. Event-based cameras present an attractive solution to this problem, mitigating the issues of traditional cameras by tracking changes in the footage on a frame-by-frame basis. In this paper, we introduce a pipeline which can be used to generate realistic synthetic data of an event-based camera mounted to an AUV (Autonomous Underwater Vehicle) in an underwater environment for training vision models. We demonstrate the effectiveness of our pipeline using the task of rock detection with poor visibility and suspended particulate matter, but the approach can be generalized to other underwater tasks.

cs.CV↗

arXiv2025

MLPerf Automotive

We present MLPerf Automotive, the first standardized public benchmark for evaluating Machine Learning systems that are deployed for AI acceleration in automotive systems. Developed through a collaborative partnership between MLCommons and the Autonomous Vehicle Computing Consortium, this benchmark addresses the need for standardized performance evaluation methodologies in automotive machine learning systems. Existing benchmark suites cannot be utilized for these systems since automotive workloads have unique constraints including safety and real-time processing that distinguish them from the domains that previously introduced benchmarks target. Our benchmarking framework provides latency and accuracy metrics along with evaluation protocols that enable consistent and reproducible performance comparisons across different hardware platforms and software implementations. The first iteration of the benchmark consists of automotive perception tasks in 2D object detection, 2D semantic segmentation, and 3D object detection. We describe the methodology behind the benchmark design including the task selection, reference models, and submission rules. We also discuss the first round of benchmark submissions and the challenges involved in acquiring the datasets and the engineering efforts to develop the reference implementations. Our benchmark code is available at https://github.com/mlcommons/mlperf_automotive.

cs.LG↗

arXiv2025

AGN Feedback Models and AGN Demographics I: Radio-Mode AGN in EAGLE, SIMBA and TNG100 are Inconsistent with Observations

We compare predictions of how Active Galactic Nuclei (AGN) populate host galaxies at low redshifts to observations, finding large discrepancies between cosmological simulation predictions and observed patterns. Modern cosmological simulations include AGN feedback models tuned to reproduce the observed galaxy stellar mass function. However, due to a lack of real understanding of the physics of AGN feedback, these models vary significantly across simulations. To distinguish between the models and potentially test the underlying physics, we carry out independent tests of these models. In an earlier study, we found that $F_{\rm AGN}$ -- the observed completeness-corrected fraction of galaxies hosting radio AGN with an Eddington ratio $\lambda > 10^{-3}$ -- to be a strong function of host galaxy stellar mass ($M_\star$) but nearly independent of host specific star formation rates (sSFR) at fixed $M_\star$. In this study, we test the radio mode AGN feedback models of the EAGLE, SIMBA, and TNG100 simulations by comparing their predictions of $F_{\rm AGN} \left(M_\star \right)$ to our observational constraint. We find that none of these simulations even qualitatively reproduce the observed dependencies of $F_{\rm AGN}$ on $M_\star$ and sSFR. Finally, we find that although the given TNG100 model could be modified in order to better reproduce the observed $F_{\rm AGN}$ trend, this modification would likely also change its prediction for the local stellar mass function and star formation rates -- key observations used for calibrating the simulation in the first place. Our findings highlight a pressing need to revisit the AGN feedback prescriptions in EAGLE, SIMBA, TNG100 and other similar models.

astro-ph.GA↗

arXiv2025

Is AI currently capable of identifying wild oysters? A comparison of human annotators against the AI model, ODYSSEE

Oysters are ecologically and commercially important species that require frequent monitoring to track population demographics (e.g. abundance, growth, mortality). Current methods of monitoring oyster reefs often require destructive sampling methods and extensive manual effort. Therefore, they are suboptimal for small-scale or sensitive environments. A recent alternative, the ODYSSEE model, was developed to use deep learning techniques to identify live oysters using video or images taken in the field of oyster reefs to assess abundance. The validity of this model in identifying live oysters on a reef was compared to expert and non-expert annotators. In addition, we identified potential sources of prediction error. Although the model can make inferences significantly faster than expert and non-expert annotators (39.6 s, $2.34 \pm 0.61$ h, $4.50 \pm 1.46$ h, respectively), the model overpredicted the number of live oysters, achieving lower accuracy (63\%) in identifying live oysters compared to experts (74\%) and non-experts (75\%) alike. Image quality was an important factor in determining the accuracy of the model and the annotators. Better quality images improved human accuracy and worsened model accuracy. Although ODYSSEE was not sufficiently accurate, we anticipate that future training on higher-quality images, utilizing additional live imagery, and incorporating additional annotation training classes will greatly improve the model's predictive power based on the results of this analysis. Future research should address methods that improve the detection of living vs. dead oysters.

cs.AI↗

arXiv2024

MLPerf Power: Benchmarking the Energy Efficiency of Machine Learning Systems from Microwatts to Megawatts for Sustainable AI

Rapid adoption of machine learning (ML) technologies has led to a surge in power consumption across diverse systems, from tiny IoT devices to massive datacenter clusters. Benchmarking the energy efficiency of these systems is crucial for optimization, but presents novel challenges due to the variety of hardware platforms, workload characteristics, and system-level interactions. This paper introduces MLPerf Power, a comprehensive benchmarking methodology with capabilities to evaluate the energy efficiency of ML systems at power levels ranging from microwatts to megawatts. Developed by a consortium of industry professionals from more than 20 organizations, MLPerf Power establishes rules and best practices to ensure comparability across diverse architectures. We use representative workloads from the MLPerf benchmark suite to collect 1,841 reproducible measurements from 60 systems across the entire range of ML deployment scales. Our analysis reveals trade-offs between performance, complexity, and energy efficiency across this wide range of systems, providing actionable insights for designing optimized ML solutions from the smallest edge devices to the largest cloud infrastructures. This work emphasizes the importance of energy efficiency as a key metric in the evaluation and comparison of the ML system, laying the foundation for future research in this critical area. We discuss the implications for developing sustainable AI solutions and standardizing energy efficiency benchmarking for ML systems.

cs.AR↗

arXiv2024

ODYSSEE: Oyster Detection Yielded by Sensor Systems on Edge Electronics

Oysters are a vital keystone species in coastal ecosystems, providing significant economic, environmental, and cultural benefits. As the importance of oysters grows, so does the relevance of autonomous systems for their detection and monitoring. However, current monitoring strategies often rely on destructive methods. While manual identification of oysters from video footage is non-destructive, it is time-consuming, requires expert input, and is further complicated by the challenges of the underwater environment. To address these challenges, we propose a novel pipeline using stable diffusion to augment a collected real dataset with realistic synthetic data. This method enhances the dataset used to train a YOLOv10-based vision model. The model is then deployed and tested on an edge platform in underwater robotics, achieving a state-of-the-art 0.657 mAP@50 for oyster detection on the Aqua2 platform.

cs.CV↗

arXiv2024

Radio AGN Activity in Low Redshift Galaxies is Not Directly Related to Star Formation Rates

We examine the demographics of radio-emitting active galactic nuclei (AGN) in the local universe as a function of host galaxy properties, most notably both stellar mass and star formation rate. Radio AGN activity is theoretically implicated in helping reduce star formation rates of galaxies, and therefore it is natural to investigate the relationship between these two galaxy properties. We use a sample of around 10, 000 galaxies from the Mapping Nearby Galaxies at APO (MaNGA) survey, part of the Sloan Digital Sky Survey IV (SDSS-IV), along with the Faint Images of the Radio Sky at Twenty centimeters (FIRST) radio survey and the National Radio Astronomy Observatory (NRAO) Very Large Array (VLA) Sky Survey (NVSS). There are 1,126 galaxies in MaNGA with radio detections. Using star formation rate and stellar mass estimates based on Pipe3D, inferred from the high signal-to-noise ratio measurements from MaNGA, we show that star formation rates are strongly correlated with 20 cm radio emission, as expected. We identify as radio AGN those radio emitters that are much stronger than expected from the star formation rate. Using this sample of AGN, the well-measured stellar velocity dispersions from MaNGA, and the black hole M-sigma relationship, we examine the Eddington ratio distribution and its dependence on stellar mass and star formation rate. We find that the Eddington ratio distribution depends strongly on stellar mass, with more massive galaxies having larger Eddington ratios. As found in previous studies, the AGN fraction increases rapidly with stellar mass. We do not find any dependence on star formation rate, specific star formation rate, or velocity dispersion when controlling for stellar mass. We conclude that galaxy star formation rates appear to be unrelated to the presence or absence of a radio AGN, which may be useful in constraining theoretical models of AGN feedback.

astro-ph.GA↗

arXiv2024

Croissant: A Metadata Format for ML-Ready Datasets

Data is a critical resource for machine learning (ML), yet working with data remains a key friction point. This paper introduces Croissant, a metadata format for datasets that creates a shared representation across ML tools, frameworks, and platforms. Croissant makes datasets more discoverable, portable, and interoperable, thereby addressing significant challenges in ML data management. Croissant is already supported by several popular dataset repositories, spanning hundreds of thousands of datasets, enabling easy loading into the most commonly-used ML frameworks, regardless of where the data is stored. Our initial evaluation by human raters shows that Croissant metadata is readable, understandable, complete, yet concise.

cs.LG↗