-
Ultralow-Power Single-Sensor-Based E-Nose System Powered by Duty Cycling and Deep Learning for Real-Time Gas Identification
Authors:
Taejung Kim,
Yonggi Kim,
Wootaek Cho,
Jong-Hyun Kwak,
Jeonghoon Cho,
Youjang Pyeon,
Jae Joon Kim,
Heungjoo Shin
Abstract:
This study presents a novel, ultralow-power single-sensor-based electronic nose (e-nose) system for real-time gas identification, distinguishing itself from conventional sensor-array-based e-nose systems whose power consumption and cost increase with the number of sensors. Our system employs a single metal oxide semiconductor (MOS) sensor built on a suspended 1D nanoheater, driven by duty cycling-…
▽ More
This study presents a novel, ultralow-power single-sensor-based electronic nose (e-nose) system for real-time gas identification, distinguishing itself from conventional sensor-array-based e-nose systems whose power consumption and cost increase with the number of sensors. Our system employs a single metal oxide semiconductor (MOS) sensor built on a suspended 1D nanoheater, driven by duty cycling-characterized by repeated pulsed power inputs. The sensor's ultrafast thermal response, enabled by its small size, effectively decouples the effects of temperature and surface charge exchange on the MOS nanomaterial's conductivity. This provides distinct sensing signals that alternate between responses coupled with and decoupled from the thermally enhanced conductivity, all within a single time domain during duty cycling. The magnitude and ratio of these dual responses vary depending on the gas type and concentration, facilitating the early-stage gas identification of five gas types within 30 s via a convolutional neural network (classification accuracy = 93.9%, concentration regression error = 19.8%). Additionally, the duty-cycling mode significantly reduces power consumption by up to 90%, lowering it to 160 $μ$W to heat the sensor to 250$^\circ$C. Manufactured using only wafer-level batch microfabrication processes, this innovative e-nose system promises the facile implementation of battery-driven, long-term, and cost-effective IoT monitoring systems.
△ Less
Submitted 25 April, 2024;
originally announced April 2024.
-
Attacking the Loop: Adversarial Attacks on Graph-based Loop Closure Detection
Authors:
Jonathan J. Y. Kim,
Martin Urschler,
Patricia J. Riddle,
Jorg S. Wicker
Abstract:
With the advancement in robotics, it is becoming increasingly common for large factories and warehouses to incorporate visual SLAM (vSLAM) enabled automated robots that operate closely next to humans. This makes any adversarial attacks on vSLAM components potentially detrimental to humans working alongside them. Loop Closure Detection (LCD) is a crucial component in vSLAM that minimizes the accumu…
▽ More
With the advancement in robotics, it is becoming increasingly common for large factories and warehouses to incorporate visual SLAM (vSLAM) enabled automated robots that operate closely next to humans. This makes any adversarial attacks on vSLAM components potentially detrimental to humans working alongside them. Loop Closure Detection (LCD) is a crucial component in vSLAM that minimizes the accumulation of drift in map**, since even a small drift can accumulate into a significant drift over time. A prior work by Kim et al., SymbioLCD2, unified visual features and semantic objects into a single graph structure for finding loop closure candidates. While this provided a performance improvement over visual feature-based LCD, it also created a single point of vulnerability for potential graph-based adversarial attacks. Unlike previously reported visual-patch based attacks, small graph perturbations are far more challenging to detect, making them a more significant threat. In this paper, we present Adversarial-LCD, a novel black-box evasion attack framework that employs an eigencentrality-based perturbation method and an SVM-RBF surrogate model with a Weisfeiler-Lehman feature extractor for attacking graph-based LCD. Our evaluation shows that the attack performance of Adversarial-LCD with the SVM-RBF surrogate model was superior to that of other machine learning surrogate algorithms, including SVM-linear, SVM-polynomial, and Bayesian classifier, demonstrating the effectiveness of our attack framework. Furthermore, we show that our eigencentrality-based perturbation method outperforms other algorithms, such as Random-walk and Shortest-path, highlighting the efficiency of Adversarial-LCD's perturbation selection method.
△ Less
Submitted 12 December, 2023;
originally announced December 2023.
-
When sound wave meets the neutrino anomaly
Authors:
Jae Jun Kim
Abstract:
We propose that the sound wave coming to the inner side of Miniboone detector could be one of the sources for our having the neutrino anomaly in the experiment. We start with presenting a rough estimate for the size of the energy associated with the sound wave coming into the detector, the size of the loss of energy associate with the save wave as it travels down to the target medium due to their…
▽ More
We propose that the sound wave coming to the inner side of Miniboone detector could be one of the sources for our having the neutrino anomaly in the experiment. We start with presenting a rough estimate for the size of the energy associated with the sound wave coming into the detector, the size of the loss of energy associate with the save wave as it travels down to the target medium due to their gravitationally interacting with the sound wave in a classical sense. After that, we describe that the neutrino anomaly could be due to the sound wave interacting with the detector material under the pressure due to the presence of the mineral oil and their producing phonon-induced electrons via a process such as the thermionic emission, which may lead more events to be identified as electron-like events in the experiment. We also address that the sound wave may scatter with the electrons produced from the electron-photon shower.
△ Less
Submitted 17 November, 2023;
originally announced November 2023.
-
When a complementarity in the neutrino and the quark mixing meets a parameter symmetry and its implications to the unitarity
Authors:
Jae Jun Kim
Abstract:
We present a self-complementarity that complements relationship among the parameters in the neutrino and the quark mixing matrix and address its implications to the unitarity of the quark mixing matrix and the mixing between the neutrino and the quark sector. First, we show how a complementarity with a phase being introduced as an extra parameter can be held in the nine independent schemes of para…
▽ More
We present a self-complementarity that complements relationship among the parameters in the neutrino and the quark mixing matrix and address its implications to the unitarity of the quark mixing matrix and the mixing between the neutrino and the quark sector. First, we show how a complementarity with a phase being introduced as an extra parameter can be held in the nine independent schemes of parameterizing the matrix introducing a discrete parameter symmetry and a combination of sine functions, a part of Jarlskog invariant, with a certain size of uncertainty and use them. After that, for the first time, we show that we can use the uncertainty associated with the complementarity as an empirical constraint complementing that among the diagonal elements in the neutrino mixing matrix and our result shows one that is comparable to that was reported before. We do the same for the quark sector and then we discuss its implication in relation to the size of the uncertainty among the elements. Our study indicates that we may need to be more cautious when testing for the unitarity of the quark mixing matrix and opens up a possibility of estimating the size of mixing between the two sectors.
△ Less
Submitted 24 May, 2024; v1 submitted 31 August, 2023;
originally announced September 2023.
-
Exploring User Perspectives on ChatGPT: Applications, Perceptions, and Implications for AI-Integrated Education
Authors:
Reza Hadi Mogavi,
Chao Deng,
Justin Juho Kim,
Pengyuan Zhou,
Young D. Kwon,
Ahmed Hosny Saleh Metwally,
Ahmed Tlili,
Simone Bassanelli,
Antonio Bucchiarone,
Sujit Gujar,
Lennart E. Nacke,
Pan Hui
Abstract:
To foster the development of pedagogically potent and ethically sound AI-integrated learning landscapes, it is pivotal to critically explore the perceptions and experiences of the users immersed in these contexts. In this study, we perform a thorough qualitative content analysis across four key social media platforms. Our goal is to understand the user experience (UX) and views of early adopters o…
▽ More
To foster the development of pedagogically potent and ethically sound AI-integrated learning landscapes, it is pivotal to critically explore the perceptions and experiences of the users immersed in these contexts. In this study, we perform a thorough qualitative content analysis across four key social media platforms. Our goal is to understand the user experience (UX) and views of early adopters of ChatGPT across different educational sectors. The results of our research show that ChatGPT is most commonly used in the domains of higher education, K-12 education, and practical skills training. In social media dialogues, the topics most frequently associated with ChatGPT are productivity, efficiency, and ethics. Early adopters' attitudes towards ChatGPT are multifaceted. On one hand, some users view it as a transformative tool capable of amplifying student self-efficacy and learning motivation. On the other hand, there is a degree of apprehension among concerned users. They worry about a potential overdependence on the AI system, which they fear might encourage superficial learning habits and erode students' social and critical thinking skills. This dichotomy of opinions underscores the complexity of Human-AI Interaction in educational contexts. Our investigation adds depth to this ongoing discourse, providing crowd-sourced insights for educators and learners who are considering incorporating ChatGPT or similar generative AI tools into their pedagogical strategies.
△ Less
Submitted 25 November, 2023; v1 submitted 22 May, 2023;
originally announced May 2023.
-
Closing the Loop: Graph Networks to Unify Semantic Objects and Visual Features for Multi-object Scenes
Authors:
Jonathan J. Y. Kim,
Martin Urschler,
Patricia J. Riddle,
Jörg S. Wicker
Abstract:
In Simultaneous Localization and Map** (SLAM), Loop Closure Detection (LCD) is essential to minimize drift when recognizing previously visited places. Visual Bag-of-Words (vBoW) has been an LCD algorithm of choice for many state-of-the-art SLAM systems. It uses a set of visual features to provide robust place recognition but fails to perceive the semantics or spatial relationship between feature…
▽ More
In Simultaneous Localization and Map** (SLAM), Loop Closure Detection (LCD) is essential to minimize drift when recognizing previously visited places. Visual Bag-of-Words (vBoW) has been an LCD algorithm of choice for many state-of-the-art SLAM systems. It uses a set of visual features to provide robust place recognition but fails to perceive the semantics or spatial relationship between feature points. Previous work has mainly focused on addressing these issues by combining vBoW with semantic and spatial information from objects in the scene. However, they are unable to exploit spatial information of local visual features and lack a structure that unifies semantic objects and visual features, therefore limiting the symbiosis between the two components. This paper proposes SymbioLCD2, which creates a unified graph structure to integrate semantic objects and visual features symbiotically. Our novel graph-based LCD system utilizes the unified graph structure by applying a Weisfeiler-Lehman graph kernel with temporal constraints to robustly predict loop closure candidates. Evaluation of the proposed system shows that having a unified graph structure incorporating semantic objects and visual features improves LCD prediction accuracy, illustrating that the proposed graph structure provides a strong symbiosis between these two complementary components. It also outperforms other Machine Learning algorithms - such as SVM, Decision Tree, Random Forest, Neural Network and GNN based Graph Matching Networks. Furthermore, it has shown good performance in detecting loop closure candidates earlier than state-of-the-art SLAM systems, demonstrating that extended semantic and spatial awareness from the unified graph structure significantly impacts LCD performance.
△ Less
Submitted 23 September, 2022;
originally announced September 2022.
-
Consideration of the Intricacies Inherent in Molecular Beam Epitaxy of the NaCl/GaAs System
Authors:
Brelon J. May,
Jae ** Kim,
Patrick Walker,
William E. McMahon,
Helio R. Moutinho,
Aaron J. Ptak,
David L. Young
Abstract:
The high cost of substrates for III-V growth can be cost limiting for technologies that require large semiconductor areas. Thus, being able to separate device layers and reuse the original substrate is highly desirable, but existing techniques to lift off a film from a substrate have substantial drawbacks. This work discusses some of the complexities with the growth of a water-soluble, alkali-hali…
▽ More
The high cost of substrates for III-V growth can be cost limiting for technologies that require large semiconductor areas. Thus, being able to separate device layers and reuse the original substrate is highly desirable, but existing techniques to lift off a film from a substrate have substantial drawbacks. This work discusses some of the complexities with the growth of a water-soluble, alkali-halide salt thin film between a III-V substrate and overlayer. Much of the difficulty stems from the growth of GaAs on an actively decomposing NaCl surface at elevated temperatures. Interestingly, the presence of an in-situ electron beam incident on the NaCl surface, prior to and during GaAs deposition, affects the crystallinity and morphology of the III-V overlayer. Here we investigate a wide range of growth temperatures and the timing of the im**ing flux of both elemental sources and high energy electrons at different points during the growth. We show that an assortment of morphologies (discrete islands, porous material, and fully dense layers with sharp interfaces) and crystallinity (amorphous, crystalline, and highly textured) occur depending on the specific growth conditions, driven largely by changes in GaAs nucleation which is greatly affected by the presence of the reflection high energy electron diffraction beam.
△ Less
Submitted 9 February, 2022;
originally announced February 2022.
-
SymbioLCD: Ensemble-Based Loop Closure Detection using CNN-Extracted Objects and Visual Bag-of-Words
Authors:
Jonathan J. Y. Kim,
Martin Urschler,
Patricia J. Riddle,
Jörg S. Wicker
Abstract:
Loop closure detection is an essential tool of Simultaneous Localization and Map** (SLAM) to minimize drift in its localization. Many state-of-the-art loop closure detection (LCD) algorithms use visual Bag-of-Words (vBoW), which is robust against partial occlusions in a scene but cannot perceive the semantics or spatial relationships between feature points. CNN object extraction can address thos…
▽ More
Loop closure detection is an essential tool of Simultaneous Localization and Map** (SLAM) to minimize drift in its localization. Many state-of-the-art loop closure detection (LCD) algorithms use visual Bag-of-Words (vBoW), which is robust against partial occlusions in a scene but cannot perceive the semantics or spatial relationships between feature points. CNN object extraction can address those issues, by providing semantic labels and spatial relationships between objects in a scene. Previous work has mainly focused on replacing vBoW with CNN-derived features. In this paper, we propose SymbioLCD, a novel ensemble-based LCD that utilizes both CNN-extracted objects and vBoW features for LCD candidate prediction. When used in tandem, the added elements of object semantics and spatial-awareness create a more robust and symbiotic loop closure detection system. The proposed SymbioLCD uses scale-invariant spatial and semantic matching, Hausdorff distance with temporal constraints, and a Random Forest that utilizes combined information from both CNN-extracted objects and vBoW features for predicting accurate loop closure candidates. Evaluation of the proposed method shows it outperforms other Machine Learning (ML) algorithms - such as SVM, Decision Tree and Neural Network, and demonstrates that there is a strong symbiosis between CNN-extracted object information and vBoW features which assists accurate LCD candidate prediction. Furthermore, it is able to perceive loop closure candidates earlier than state-of-the-art SLAM algorithms, utilizing added spatial and semantic information from CNN-extracted objects.
△ Less
Submitted 21 October, 2021;
originally announced October 2021.
-
PHANGS-ALMA Data Processing and Pipeline
Authors:
Adam K. Leroy,
Annie Hughes,
Daizhong Liu,
Jerome Pety,
Erik Rosolowsky,
Toshiki Saito,
Eva Schinnerer,
Andreas Schruba,
Antonio Usero,
Christopher M. Faesi,
Cinthya N. Herrera,
Melanie Chevance,
Alexander P. S. Hygate,
Amanda A. Kepley,
Eric W. Koch,
Miguel Querejeta,
Kazimierz Sliwa,
David Will,
Christine D. Wilson,
Gagandeep S. Anand,
Ashley Barnes,
Francesco Belfiore,
Ivana Beslic,
Frank Bigiel,
Guillermo A. Blanc
, et al. (43 additional authors not shown)
Abstract:
We describe the processing of the PHANGS-ALMA survey and present the PHANGS-ALMA pipeline, a public software package that processes calibrated interferometric and total power data into science-ready data products. PHANGS-ALMA is a large, high-resolution survey of CO J=2-1 emission from nearby galaxies. The observations combine ALMA's main 12-m array, the 7-m array, and total power observations and…
▽ More
We describe the processing of the PHANGS-ALMA survey and present the PHANGS-ALMA pipeline, a public software package that processes calibrated interferometric and total power data into science-ready data products. PHANGS-ALMA is a large, high-resolution survey of CO J=2-1 emission from nearby galaxies. The observations combine ALMA's main 12-m array, the 7-m array, and total power observations and use mosaics of dozens to hundreds of individual pointings. We describe the processing of the u-v data, imaging and deconvolution, linear mosaicking, combining interferometer and total power data, noise estimation, masking, data product creation, and quality assurance. Our pipeline has a general design and can also be applied to VLA and ALMA observations of other spectral lines and continuum emission. We highlight our recipe for deconvolution of complex spectral line observations, which combines multiscale clean, single scale clean, and automatic mask generation in a way that appears robust and effective. We also emphasize our two-track approach to masking and data product creation. We construct one set of "broadly masked" data products, which have high completeness but significant contamination by noise, and another set of "strictly masked" data products, which have high confidence but exclude faint, low signal-to-noise emission. Our quality assurance tests, supported by simulations, demonstrate that 12-m+7-m deconvolved data recover a total flux that is significantly closer to the total power flux than the 7-m deconvolved data alone. In the appendices, we measure the stability of the ALMA total power calibration in PHANGS--ALMA and test the performance of popular short-spacing correction algorithms.
△ Less
Submitted 14 April, 2021;
originally announced April 2021.
-
Expressive Text-to-Speech using Style Tag
Authors:
Minchan Kim,
Sung Jun Cheon,
Byoung ** Choi,
Jong ** Kim,
Nam Soo Kim
Abstract:
As recent text-to-speech (TTS) systems have been rapidly improved in speech quality and generation speed, many researchers now focus on a more challenging issue: expressive TTS. To control speaking styles, existing expressive TTS models use categorical style index or reference speech as style input. In this work, we propose StyleTagging-TTS (ST-TTS), a novel expressive TTS model that utilizes a st…
▽ More
As recent text-to-speech (TTS) systems have been rapidly improved in speech quality and generation speed, many researchers now focus on a more challenging issue: expressive TTS. To control speaking styles, existing expressive TTS models use categorical style index or reference speech as style input. In this work, we propose StyleTagging-TTS (ST-TTS), a novel expressive TTS model that utilizes a style tag written in natural language. Using a style-tagged TTS dataset and a pre-trained language model, we modeled the relationship between linguistic embedding and speaking style domain, which enables our model to work even with style tags unseen during training. As style tag is written in natural language, it can control speaking style in a more intuitive, interpretable, and scalable way compared with style index or reference speech. In addition, in terms of model architecture, we propose an efficient non-autoregressive (NAR) TTS architecture with single-stage training. The experimental result shows that ST-TTS outperforms the existing expressive TTS model, Tacotron2-GST in speech quality and expressiveness.
△ Less
Submitted 6 October, 2022; v1 submitted 1 April, 2021;
originally announced April 2021.
-
Multi-Block Sparse Functional Principal Components Analysis for Longitudinal Microbiome Multi-Omics Data
Authors:
Ling**g Jiang,
Chris Elrod,
Jane J. Kim,
Austin D. Swafford,
Rob Knight,
Wesley K. Thompson
Abstract:
Microbiome researchers often need to model the temporal dynamics of multiple complex, nonlinear outcome trajectories simultaneously. This motivates our development of multivariate Sparse Functional Principal Components Analysis (mSFPCA), extending existing SFPCA methods to simultaneously characterize multiple temporal trajectories and their inter-relationships. As with existing SFPCA methods, the…
▽ More
Microbiome researchers often need to model the temporal dynamics of multiple complex, nonlinear outcome trajectories simultaneously. This motivates our development of multivariate Sparse Functional Principal Components Analysis (mSFPCA), extending existing SFPCA methods to simultaneously characterize multiple temporal trajectories and their inter-relationships. As with existing SFPCA methods, the mSFPCA algorithm characterizes each trajectory as a smooth mean plus a weighted combination of the smooth major modes of variation about the mean, where the weights are given by the component scores for each subject. Unlike existing SFPCA methods, the mSFPCA algorithm allows estimation of multiple trajectories simultaneously, such that the component scores, which are constrained to be independent within a particular outcome for identifiability, may be arbitrarily correlated with component scores for other outcomes. A Cholesky decomposition is used to estimate the component score covariance matrix efficiently and guarantee positive semi-definiteness given these constraints. Mutual information is used to assess the strength of marginal and conditional temporal associations across outcome trajectories. Importantly, we implement mSFPCA as a Bayesian algorithm using R and stan, enabling easy use of packages such as PSIS-LOO for model selection and graphical posterior predictive checks to assess the validity of mSFPCA models. Although we focus on application of mSFPCA to microbiome data in this paper, the mSFPCA model is of general utility and can be used in a wide range of real-world applications.
△ Less
Submitted 5 February, 2021; v1 submitted 29 January, 2021;
originally announced February 2021.
-
Difference in chemical composition between the bright and faint red clump stars in the Milky Way bulge
Authors:
Dongwook Lim,
Young-Wook Lee,
Andreas Koch,
Seungsoo Hong,
Christian I. Johnson,
Jenny J. Kim,
Chul Chung,
Mario Mateo,
John I. Bailey, III
Abstract:
The double red clump (RC) observed in color-magnitude diagrams of the Milky Way bulge is at the heart of the current debate on the structure and formation origin of the bulge. This feature can be explained by the difference between the two RCs either in distance ("X-shaped scenario") or in chemical composition ("multiple-population scenario"). Here we report our high-resolution spectroscopy for th…
▽ More
The double red clump (RC) observed in color-magnitude diagrams of the Milky Way bulge is at the heart of the current debate on the structure and formation origin of the bulge. This feature can be explained by the difference between the two RCs either in distance ("X-shaped scenario") or in chemical composition ("multiple-population scenario"). Here we report our high-resolution spectroscopy for the RC and red giant branch stars in a high-latitude field (b ~ -8.5$°$) of the bulge. We find a difference in [Fe/H] between the stars in the bright and faint RC regimes, in the sense that the bright stars are enhanced in [Fe/H] with respect to the faint stars by 0.149 $\pm$ 0.036 dex. The stars on the bright RC are also enhanced in [Na/Fe] but appear to be depleted in [Al/Fe] and [O/Fe], although more observations are required to confirm the significance of these differences. Interestingly, these chemical patterns are similar to those observed among multiple stellar populations in the metal-rich bulge globular cluster Terzan 5. In addition, we find a number of Na-rich stars, which would corroborate the presence of multiple populations in the bulge. Our results support an origin of the double RC from dissolved globular clusters that harbor multiple stellar populations. Thus, our study suggests that a substantial fraction of the outer bulge stars would have originated from the assembly of such stellar systems in the early phase of the Milky Way formation.
△ Less
Submitted 7 December, 2020;
originally announced December 2020.
-
Pre-supernova feedback mechanisms drive the destruction of molecular clouds in nearby star-forming disc galaxies
Authors:
Mélanie Chevance,
J. M. Diederik Kruijssen,
Mark R. Krumholz,
Brent Groves,
Benjamin W. Keller,
Annie Hughes,
Simon C. O. Glover,
Jonathan D. Henshaw,
Cinthya N. Herrera,
Jenny J. Kim,
Adam K. Leroy,
Jérôme Pety,
Alessandro Razza,
Erik Rosolowsky,
Eva Schinnerer,
Andreas Schruba,
Ashley T. Barnes,
Frank Bigiel,
Guillermo A. Blanc,
Eric Emsellem,
Christopher M. Faesi,
Kathryn Grasha,
Ralf S. Klessen,
Kathryn Kreckel,
Daizhong Liu
, et al. (6 additional authors not shown)
Abstract:
It is a major open question which physical processes stop the accretion of gas onto giant molecular clouds (GMCs) and limit the efficiency at which gas is converted into stars within these GMCs. While feedback from supernova explosions has been the popular feedback mechanism included in simulations of galaxy formation and evolution, `early' feedback mechanisms such as stellar winds, photoionisatio…
▽ More
It is a major open question which physical processes stop the accretion of gas onto giant molecular clouds (GMCs) and limit the efficiency at which gas is converted into stars within these GMCs. While feedback from supernova explosions has been the popular feedback mechanism included in simulations of galaxy formation and evolution, `early' feedback mechanisms such as stellar winds, photoionisation and radiation pressure are expected to play an important role in dispersing the gas after the onset of star formation. These feedback processes typically take place on small scales ($\sim 10-100$ pc) and their effects have therefore been difficult to constrain in environments other than the Milky Way. We apply a novel statistical method to $\sim 1$" resolution maps of CO and Ha emission across a sample of nine nearby disc galaxies, in order to measure the time over which GMCs are dispersed by feedback from young, high-mass stars, as a function of the galactic environment. We find that GMCs are typically dispersed within $\sim$ 3 Myr after the emergence of unembedded high-mass stars, showing no significant trend with galactocentric radius. Comparison with analytical predictions demonstrates that, independently of the environment, early feedback mechanisms (particularly photoionisation and stellar winds) play a crucial role in dispersing GMCs and limiting their star formation efficiency in nearby galaxies. Finally, we show that the efficiency at which the energy injected by these early feedback mechanisms couples with the parent GMC is relatively low (a few tens of per cent), such that the vast majority of momentum and energy emitted by the young stellar populations escapes the parent GMC.
△ Less
Submitted 26 October, 2020;
originally announced October 2020.
-
Measuring the mixing scale of the ISM within nearby spiral galaxies
Authors:
Kathryn Kreckel,
I-Ting Ho,
Guillermo A. Blanc,
Simon C. O. Glover,
Brent Groves,
Erik Rosolowsky,
Frank Bigiel,
Mederic Boquien,
Melanie Chevance,
Daniel A. Dale,
Sinan Deger,
Eric Emsellem,
Kathryn Grasha,
Jenny J. Kim,
Ralf S. Klessen,
J. M. Diederik Kruijssen,
Janice C. Lee,
Adam K. Leroy,
Daizhong Liu,
Rebecca McElroy,
Sharon E. Meidt,
Ismael Pessa,
Patricia Sanchez-Blazquez,
Karin Sandstrom,
Francesco Santoro
, et al. (6 additional authors not shown)
Abstract:
The spatial distribution of metals reflects, and can be used to constrain, the processes of chemical enrichment and mixing. Using PHANGS-MUSE optical integral field spectroscopy, we measure the gas phase oxygen abundances (metallicities) across 7,138 HII regions in a sample of eight nearby disc galaxies. In Paper I (Kreckel et al. 2019) we measure and report linear radial gradients in the metallic…
▽ More
The spatial distribution of metals reflects, and can be used to constrain, the processes of chemical enrichment and mixing. Using PHANGS-MUSE optical integral field spectroscopy, we measure the gas phase oxygen abundances (metallicities) across 7,138 HII regions in a sample of eight nearby disc galaxies. In Paper I (Kreckel et al. 2019) we measure and report linear radial gradients in the metallicities of each galaxy, and qualitatively searched for azimuthal abundance variations. Here, we examine the two-dimensional variation in abundances once the radial gradient is subtracted, Delta(O/H), in order to quantify the homogeneity of the metal distribution and to measure the mixing scale over which HII region metallicities are correlated. We observe low (0.03--0.05 dex) scatter in Delta(O/H) globally in all galaxies, with significantly lower (0.02--0.03 dex) scatter on small (<600 pc) spatial scales. This is consistent with the measurement uncertainties, and implies the two-dimensional metallicity distribution is highly correlated on scales of <600 pc. We compute the two point correlation function for metals in the disc in order to quantify the scale lengths associated with the observed homogeneity. This mixing scale is observed to correlate better with the local gas velocity dispersion (of both cold and ionized gas) than with the star formation rate. Selecting only HII regions with enhanced abundances relative to a linear radial gradient, we do not observe increased homogeneity on small scales. This suggests that the observed homogeneity is driven by the mixing introducing material from large scales rather than by pollution from recent and on-going star formation.
△ Less
Submitted 4 September, 2020;
originally announced September 2020.
-
Molecular Gas Properties on Cloud Scales Across the Local Star-forming Galaxy Population
Authors:
Jiayi Sun,
Adam K. Leroy,
Eva Schinnerer,
Annie Hughes,
Erik Rosolowsky,
Miguel Querejeta,
Andreas Schruba,
Daizhong Liu,
Toshiki Saito,
Cinthya N. Herrera,
Christopher Faesi,
Antonio Usero,
Jérôme Pety,
J. M. Diederik Kruijssen,
Eve C. Ostriker,
Frank Bigiel,
Guillermo A. Blanc,
Alberto D. Bolatto,
Médéric Boquien,
Mélanie Chevance,
Daniel A. Dale,
Sinan Deger,
Eric Emsellem,
Simon C. O. Glover,
Kathryn Grasha
, et al. (12 additional authors not shown)
Abstract:
Using the PHANGS-ALMA CO (2-1) survey, we characterize molecular gas properties on ${\sim}$100 pc scales across 102,778 independent sightlines in 70 nearby galaxies. This yields the best synthetic view of molecular gas properties on cloud scales across the local star-forming galaxy population obtained to date. Consistent with previous studies, we observe a wide range of molecular gas surface densi…
▽ More
Using the PHANGS-ALMA CO (2-1) survey, we characterize molecular gas properties on ${\sim}$100 pc scales across 102,778 independent sightlines in 70 nearby galaxies. This yields the best synthetic view of molecular gas properties on cloud scales across the local star-forming galaxy population obtained to date. Consistent with previous studies, we observe a wide range of molecular gas surface densities (3.4 dex), velocity dispersions (1.7 dex), and turbulent pressures (6.5 dex) across the galaxies in our sample. Under simplifying assumptions about sub-resolution gas structure, the inferred virial parameters suggest that the kinetic energy of the molecular gas typically exceeds its self-gravitational binding energy at ${\sim}$100 pc scales by a modest factor (1.3 on average). We find that the cloud-scale surface density, velocity dispersion, and turbulent pressure (1) increase towards the inner parts of galaxies, (2) are exceptionally high in the centers of barred galaxies (where the gas also appears less gravitationally bound), and (3) are moderately higher in spiral arms than in inter-arm regions. The galaxy-wide averages of these gas properties also correlate with the integrated stellar mass, star formation rate, and offset from the star-forming main sequence of the host galaxies. These correlations persist even when we exclude regions with extraordinary gas properties in galaxy centers, which contribute significantly to the inter-galaxy variations. Our results provide key empirical constraints on the physical link between molecular cloud populations and their galactic environment.
△ Less
Submitted 3 September, 2020;
originally announced September 2020.
-
Multiple populations in globular clusters: Unified efforts from stellar evolution and chemical evolution models
Authors:
Sohee Jang,
Jenny J. Kim,
Young-Wook Lee
Abstract:
Recent stellar evolution models for globular clusters (GCs) in multiple population paradigm suggest that horizontal-branch (HB) morphology and mean period of type ab RR Lyrae variables are mostly determined by He & CNO abundances and relative ages for subpopulations. These parameters are also provided by chemical evolution models constructed to reproduce the Na-O anti-correlation. Therefore, a con…
▽ More
Recent stellar evolution models for globular clusters (GCs) in multiple population paradigm suggest that horizontal-branch (HB) morphology and mean period of type ab RR Lyrae variables are mostly determined by He & CNO abundances and relative ages for subpopulations. These parameters are also provided by chemical evolution models constructed to reproduce the Na-O anti-correlation. Therefore, a consistency check is possible between the synthetic HB and chemical evolution models. Furthermore, by combining them, a better constraint might be attained for star formation history and chemical abundances of subpopulations in GCs. We find, from such efforts made for four GCs, M4, M5, M15, and M80, that consistent results can be obtained from these two independent studies. In our unified model, He and Na abundances gradually increase over the generation, and therefore, the various extensions observed in both HB morphology and Na-O chemical pattern depend on the presence of later generation stars after the second generation. It is schematically shown that this observed diversity, however, would not be naturally explained by the models requiring dilution. Further spectroscopic observations are required, for metal-poor GCs in particular, to obtain a more detailed constraint from this approach.
△ Less
Submitted 22 October, 2019;
originally announced October 2019.
-
The Globular Cluster Origin of the Milky Way Outer Bulge: Evidence from Sodium Bimodality
Authors:
Young-Wook Lee,
Jenny J. Kim,
Christian I. Johnson,
Chul Chung,
Sohee Jang,
Dongwook Lim,
Yijung Kang
Abstract:
Recent investigations of the double red clump in the color-magnitude diagram of the Milky Way bulge cast serious doubts on the structure and formation origin of the outer bulge. Unlike previous interpretation based on an X-shaped bulge, stellar evolution models and CN-band observations have suggested that this feature is another manifestation of the multiple stellar population phenomenon observed…
▽ More
Recent investigations of the double red clump in the color-magnitude diagram of the Milky Way bulge cast serious doubts on the structure and formation origin of the outer bulge. Unlike previous interpretation based on an X-shaped bulge, stellar evolution models and CN-band observations have suggested that this feature is another manifestation of the multiple stellar population phenomenon observed in globular clusters (GCs). This new scenario requires a significant fraction of the outer bulge stars with chemical patterns uniquely observed in GCs. Here we show from homogeneous high-quality spectroscopic data that the red giant branch stars in the outer bulge (> 5.5 degrees from the Galactic center) are clearly divided into two groups according to Na abundance in the [Na/Fe] - [Fe/H] plane. The Na-rich stars are also enhanced in Al, while the differences in O and Mg are not observed between the two Na groups. The population ratio and the Na and Al differences between the two groups are also comparable with those observed in metal-rich GCs. The only plausible explanation for these chemical patterns and characteristics appears to be that the outer bulge was mostly assembled from disrupted proto-GCs in the early history of the Milky Way.
△ Less
Submitted 21 May, 2019;
originally announced May 2019.
-
One-Shot Item Search with Multimodal Data
Authors:
Jonghwa Yim,
Junghun James Kim,
Daekyu Shin
Abstract:
In the task of near similar image search, features from Deep Neural Network is often used to compare images and measure similarity. In the past, we only focused visual search in image dataset without text data. However, since deep neural network emerged, the performance of visual search becomes high enough to apply it in many industries from 3D data to multimodal data. Compared to the needs of mul…
▽ More
In the task of near similar image search, features from Deep Neural Network is often used to compare images and measure similarity. In the past, we only focused visual search in image dataset without text data. However, since deep neural network emerged, the performance of visual search becomes high enough to apply it in many industries from 3D data to multimodal data. Compared to the needs of multimodal search, there has not been sufficient researches.
In this paper, we present a method of near similar search with image and text multimodal dataset. Earlier time, similar image search, especially when searching shop** items, treated image and text separately to search similar items and reorder the results. This regards two tasks of image search and text matching as two different tasks. Our method, however, explore the vast data to compute k-nearest neighbors using both image and text.
In our experiment of similar item search, our system using multimodal data shows better performance than single data while it only increases minute computing time. For the experiment, we collected more than 15 million of accessory and six million of digital product items from online shop** websites, in which the product item comprises item images, titles, categories, and descriptions. Then we compare the performance of multimodal searching to single space searching in these datasets.
△ Less
Submitted 28 November, 2018; v1 submitted 27 November, 2018;
originally announced November 2018.
-
Explaining the multiple populations in globular clusters by multiple episodes of star formation and enrichment without gas expulsion from massive star feedback
Authors:
Jenny J. Kim,
Young-Wook Lee
Abstract:
In order to investigate the origin of multiple stellar populations found in globular clusters (GCs) in the halo and bulge of the Milky Way, we have constructed chemical evolution models for their putative low-mass progenitors. In light of recent theoretical developments, we assume that supernova blast waves undergo blowout without expelling the pre-enriched ambient gas, while relatively slow winds…
▽ More
In order to investigate the origin of multiple stellar populations found in globular clusters (GCs) in the halo and bulge of the Milky Way, we have constructed chemical evolution models for their putative low-mass progenitors. In light of recent theoretical developments, we assume that supernova blast waves undergo blowout without expelling the pre-enriched ambient gas, while relatively slow winds of massive stars, together with the winds and ejecta from low to high mass asymptotic-giant-branch stars, are all locally retained in these less massive systems. Interestingly, we find that the observed Na-O anti-correlations in metal-poor GCs can be reproduced when multiple episodes of starburst and enrichment are allowed to continue in these subsystems. A specific form of star formation history with decreasing time intervals between the successive stellar generations, however, is required to obtain this result, which is in good agreement with the parameters obtained from synthetic horizontal-branch models. The "mass budget problem" is also much alleviated by our models without ad-hoc assumptions on star formation efficiency, initial mass function, and the preferential loss of first-generation stars. We also applied these models to investigate the origin of super-He-rich red clump stars in the metal-rich bulge suggested by Lee et al. (2015). We find that chemical enrichment by the winds of massive stars can naturally reproduce the required strong He enhancement in metal-rich subsystems. Our results further underscore that gas expulsion or retention is a key factor in understanding the multiple populations in GCs.
△ Less
Submitted 1 November, 2018; v1 submitted 3 July, 2018;
originally announced July 2018.
-
Assembling the Milky Way bulge from globular clusters: Evidence from the double red clump
Authors:
Young-Wook Lee,
Seungsoo Hong,
Dongwook Lim,
Chul Chung,
Sohee Jang,
Jenny J. Kim,
Seok-Joo Joo
Abstract:
The two red clumps (RCs) observed in the color-magnitude diagram of the Milky Way bulge is widely accepted as evidence for an X-shaped structure originated from the bar instability. A drastically different interpretation has been suggested, however, based on the He-enhanced multiple stellar population phenomenon as is observed in globular clusters (GCs). Because these two scenarios imply very diff…
▽ More
The two red clumps (RCs) observed in the color-magnitude diagram of the Milky Way bulge is widely accepted as evidence for an X-shaped structure originated from the bar instability. A drastically different interpretation has been suggested, however, based on the He-enhanced multiple stellar population phenomenon as is observed in globular clusters (GCs). Because these two scenarios imply very different pictures on the formation of the bulge and elliptical galaxies, understanding the origin of the double RC is of crucial importance. Here we report our discovery that the stars in the two RCs show a significant (> 5.3 σ) difference in CN-band strength, in stark contrast to that expected in the X-shaped bulge scenario. The difference in CN abundance and the population ratio between the two RCs are comparable to those observed in GCs between the first- and later generation stars. Since CN-strong stars trace a population with enhanced N, Na, and He abundances originated in GCs, this is direct evidence that the double RC is due to the multiple population phenomenon, and that a significant population of stars in the Milky Way bulge were assembled from disrupted proto-GCs. Our result also calls for the major revision of the 3D structure of the Milky Way bulge given that the current view is based on the previous interpretation of the double RC phenomenon.
△ Less
Submitted 29 June, 2018;
originally announced July 2018.
-
Optimal conditional expectation at the video poker game Jacks or Better
Authors:
S. N. Ethier,
John Jungtae Kim,
Jiyeon Lee
Abstract:
There are 134,459 distinct initial hands at the video poker game Jacks or Better, taking suit exchangeability into account. A computer program can determine the optimal strategy (i.e., which cards to hold) for each such hand, but a complete list of these strategies would require a book-length manuscript. Instead, a hand-rank table, which fits on a single page and reproduces the optimal strategy pe…
▽ More
There are 134,459 distinct initial hands at the video poker game Jacks or Better, taking suit exchangeability into account. A computer program can determine the optimal strategy (i.e., which cards to hold) for each such hand, but a complete list of these strategies would require a book-length manuscript. Instead, a hand-rank table, which fits on a single page and reproduces the optimal strategy perfectly, was found for Jacks or Better as early as the mid 1990s. Is there a systematic way to derive such a hand-rank table? We show that there is indeed, and it involves finding the exact optimal conditional expected return, given the initial hand. In the case of Jacks or Better (paying 800, 50, 25, 9, 6, 4, 3, 2, 1, 0), this is a random variable with 1,153 distinct values, of which 766 correspond to garbage hands for which it is optimal to draw five new cards. We describe the hands corresponding to each of the remaining 387 values of the optimal conditional expected return (sorted from largest to smallest) and show how this leads readily to an optimal strategy hand-rank table for Jacks or Better. Of course, the method applies to other video poker games as well.
△ Less
Submitted 12 February, 2016;
originally announced February 2016.
-
Do** Dependent Changes in Nitrogen 2$p$ States in the Diluted Magnetic Semiconductor Ga$_{1-x}$Cr$_{x}$N
Authors:
T. Takeuchi,
Y. Harada,
T. Tokushima,
M. Taguchi,
Y. Takata,
A. Chainani,
J. J. Kim,
H. Makino,
T. Yao,
T. Yamamoto,
T. Tsukamoto,
S. Shin,
K. Kobayashi
Abstract:
We study the electronic structure of the recently discovered diluted magnetic semiconductor Ga$_{1-x}$Cr$_{x}$N ($x$ = 0.01-0.10). A systematic study of the changes in the $occupied$ and $unoccupied$ ligand (N) partial density of states (DOS) of the host lattice is carried out using N 1$s$ soft x-ray emission and absorption spectroscopy, respectively. X-ray absorption measurements confirm the wu…
▽ More
We study the electronic structure of the recently discovered diluted magnetic semiconductor Ga$_{1-x}$Cr$_{x}$N ($x$ = 0.01-0.10). A systematic study of the changes in the $occupied$ and $unoccupied$ ligand (N) partial density of states (DOS) of the host lattice is carried out using N 1$s$ soft x-ray emission and absorption spectroscopy, respectively. X-ray absorption measurements confirm the wurtzite N 2$p$ DOS and substitutional do** of Cr into Ga-sites. Coupled changes in the $occupied$ and $unoccupied$ N 2$p$ character DOS of Ga$_{1-x}$Cr$_{x}$N identify states responsible for ferromagnetism consistent with band structure calculations.
△ Less
Submitted 30 September, 2004;
originally announced September 2004.
