Search

We investigate AGN feedback from an intermediate-mass black hole at the center of a dwarf spheroidal galaxy, by performing isolated galaxy simulations using a modified version of the GADGET-3 code. We consider Leo II (PGC 34176) in the Local Group as our simulation reference model. Beginning with black hole seeds ranging from $10^3$ to $10^6$ M$_{\odot}$, our simulations focus on comparing stellar-only feedback with AGN+stellar/SN feedback over 13.7 Gyr of galactic evolution. Our results indicate that a low-mass AGN in a dwarf galaxy influences the star formation history under specific physical conditions. While AGN feedback is generally negative on star formation, instances of positive feedback were also identified. Despite measurable effects on the evolution of the dwarf host galaxy, black hole seeds exhibited only marginal growth. We tested several physical scenarios as modified models in our simulations, primarily concerning the dynamics of the central black holes, which may wander within dwarf galaxies rather than being centrally located. However, none of these adjustments significantly impacted the growth of the black hole seeds. This suggests that intermediate-mass black holes may struggle to achieve higher masses in isolated environments, with mergers and interactions likely playing crucial roles in their growth. Nevertheless, AGN feedback exhibited non-negligible effects in our simulated dwarf spheroidal galaxies, despite the assumed dominant role of stellar feedback in the low-mass regime., Comment: To be published in The Astrophysical Journal. 34 pages, 20 figures

Reuse distance analysis is a widely recognized method for application characterization that illustrates cache locality. Although there are various techniques to calculate the reuse profile from dynamic memory traces, it is both time and space-consuming due to the requirement to collect dynamic memory traces at runtime. In contrast, static analysis reuse profile estimation is a promisingly faster approach since it is calculated at compile time without running the program or collecting memory traces. This work presents a static analysis technique to estimate the reuse profile of loop-based programs. For an input program, we generate a basic block-level control flow graph and the execution count by analyzing the LLVM IR of the program. We present the memory accesses of the application kernel in a compact bracketed format and use a recursive algorithm to predict the reuse distance histogram. We deploy a separate predictor that unrolls the loop(s) for smaller bounds and generates a temporary reuse distance profile for those small cases. Using these smaller profiles, the reuse profile is extrapolated for the actual loop bound(s). We use this reuse profile to predict the cache hit rate. Results show that our model can predict cache hit rates with an average accuracy of 95% relative to the dynamic reuse profile methods., Comment: Accepted in The International Symposium on Memory Systems (MEMSYS 24), September 30 to October 03, 2024, Washington, DC, USA

Large Language Models (LLMs) have demonstrated immense potential in artificial intelligence across various domains, including healthcare. However, their efficacy is hindered by the need for high-quality labeled data, which is often expensive and time-consuming to create, particularly in low-resource domains like healthcare. To address these challenges, we propose a crowdsourcing (CS) framework enriched with quality control measures at the pre-, real-time-, and post-data gathering stages. Our study evaluated the effectiveness of enhancing data quality through its impact on LLMs (Bio-BERT) for predicting autism-related symptoms. The results show that real-time quality control improves data quality by 19 percent compared to pre-quality control. Fine-tuning Bio-BERT using crowdsourced data generally increased recall compared to the Bio-BERT baseline but lowered precision. Our findings highlighted the potential of crowdsourcing and quality control in resource-constrained environments and offered insights into optimizing healthcare LLMs for informed decision-making and improved patient care., Comment: Published in AMIA Summit, Boston, 2024. https://knowledge.amia.org/Info2024/pdf/Info2024a022/Info2024fl021

A fundamental question regarding the evolution of dwarf spheroidal galaxies is the identification of the key physical mechanisms responsible for gas depletion. Here, we focus on the study of stellar feedback in isolated dwarf spheroidal galaxies, by performing numerical simulations using a modified version of the SPH code GADGET-3. The Milky Way satellite Leo II (PGC 34176) in the Local Group was considered as our default model dwarf galaxy. The parameter space for the stellar feedback models was explored to match observational constraints of Leo II, such as residual gas mass, total mass within the tidal radius, star formation history, final stellar mass, stellar ages and metallicity. Additionally, we examined the impact of the binary fraction of stars, initial mass function, dark matter halo mass and initial gas reservoir. Many simulations revealed recent star formation quenching due to stellar feedback. In general, the gas depletion, expected star formation history, total mass of stars and total mass within the tidal radius were adequately reproduced in the simulations when compared to observational estimates. However, there were discrepancies in the distribution of stellar ages and metallicities, which suggested that the cosmic gas infall would play a more complex role in our dwarf spheroidal galaxy than captured by a monolithic infall scenario. Our results suggest that currently quenched dwarf galaxies may not necessarily need to evolve within clusters or groups, and that stellar feedback alone could be a sufficient factor in shaping at least some of these galaxies as we observe them today., Comment: To be published in The Astrophysical Journal. 33 pages, 20 figures

Health literacy is crucial to supporting good health and is a major national goal. Audio delivery of information is becoming more popular for informing oneself. In this study, we evaluate the effect of audio enhancements in the form of information emphasis and pauses with health texts of varying difficulty and we measure health information comprehension and retention. We produced audio snippets from difficult and easy text and conducted the study on Amazon Mechanical Turk (AMT). Our findings suggest that emphasis matters for both information comprehension and retention. When there is no added pause, emphasizing significant information can lower the perceived difficulty for difficult and easy texts. Comprehension is higher (54%) with correctly placed emphasis for the difficult texts compared to not adding emphasis (50%). Adding a pause lowers perceived difficulty and can improve retention but adversely affects information comprehension., Comment: This manuscript is accepted to American Medical Informatics Association summit, 2024

CMOS VLSI technology is the most dominant integration methodology prevailing in the world today. Various signal-processing blocks are made using analog or digital design techniques in MOS VLSI. An important component is the Memory unit used to store data. In the project a memory cell has been built up using analog design method. A capacitor is used as the basic storage device. The main idea behind analog memory is that the analog value of the charge or voltage stored in the capacitor is the data stored. So the dielectric quality of the capacitor becomes important here to determine how effectively it can store some charge. Analog memory is a trade off between hardware cost, chip area and accuracy or quality of storage. The circuit of analog memory cell was developed starting from the idea that required voltage will be stored in a capacitor and MOS transistors were used as switches. A given technology of integration was used and hence the dielectric property of the capacitor was fixed. By suitable circuit configuration the analog voltage value was written to the capacitor, read out when required and the charge loss was also refreshed. The results obtained are as given in the thesis., Comment: Bachelors Thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Technology (Honours) in Electronics and Electrical Communication Engineering at the Indian Institute of Technology - Kharagpur (year 2002), by Paramita Barai under the guidance of Prof. Dr. A. S. Dhar

Profiling various application characteristics, including the number of different arithmetic operations performed, memory footprint, etc., dynamically is time- and space-consuming. On the other hand, static analysis methods, although fast, can be less accurate. This paper presents an LLVM-based probabilistic static analysis method that accurately predicts different program characteristics and estimates the reuse distance profile of a program by analyzing the LLVM IR file in constant time, regardless of program input size. We generate the basic-block-level control flow graph of the target application kernel and determine basic-block execution counts by solving the linear balance equation involving the adjacent basic blocks' transition probabilities. Finally, we represent the kernel memory accesses in a bracketed format and employ a recursive algorithm to calculate the reuse distance profile. The results show that our approach can predict application characteristics accurately compared to another LLVM-based dynamic code analysis tool, Byfl., Comment: This paper was accepted at the MEMSYS '23 conference, The International Symposium on Memory Systems, October 02, 2023 - October 05, 2023, Alexandria, VA

The global need for effective disease diagnosis remains substantial, given the complexities of various disease mechanisms and diverse patient symptoms. To tackle these challenges, researchers, physicians, and patients are turning to machine learning (ML), an artificial intelligence (AI) discipline, to develop solutions. By leveraging sophisticated ML and AI methods, healthcare stakeholders gain enhanced diagnostic and treatment capabilities. However, there is a scarcity of research focused on ML algorithms for enhancing the accuracy and computational efficiency. This research investigates the capacity of machine learning algorithms to improve the transmission of heart rate data in time series healthcare metrics, concentrating particularly on optimizing accuracy and efficiency. By exploring various ML algorithms used in healthcare applications, the review presents the latest trends and approaches in ML-based disease diagnosis (MLBDD). The factors under consideration include the algorithm utilized, the types of diseases targeted, the data types employed, the applications, and the evaluation metrics. This review aims to shed light on the prospects of ML in healthcare, particularly in disease diagnosis. By analyzing the current literature, the study provides insights into state-of-the-art methodologies and their performance metrics., Comment: 8 pages

We investigate the coalescence of massive black hole ($M_{\rm BH}\gtrsim 10^{6}~\rm M_{\odot}$) binaries (MBHBs) at $63 \sigma$) overdense regions ($M_h\sim 10^{12}~\rm M_{\odot}$ dark matter halos at $z = 6$) of the Universe. We first analyse the impact of different resolutions and AGN feedback prescriptions on the merger rate, assuming instantaneous mergers. Then, we compute the halo bias correction factor due to the overdense simulated region. Our simulations predict merger rates that range between 3 - 15 $\rm yr^{-1}$ at $z\sim 6$, depending on the run considered, and after correcting for a bias factor of $\sim 20-30$. For our fiducial model, we further consider the effect of delay in the MBHB coalescence due to dynamical friction. We find that 83 per cent of MBHBs will merge within the Hubble time, and 21 per cent within 1 Gyr, namely the age of the Universe at $z > 6$. We finally compute the expected properties of the gravitational wave (GW) signals and find the fraction of LISA detectable events with high signal-to-noise ratio (SNR $>$ 5) to range between 66-69 per cent. However, identifying the electro-magnetic counterpart of these events remains challenging due to the poor LISA sky localization that, for the loudest signals ($\mathcal M_c\sim 10^6~\rm M_{\odot}$ at $z=6$), is around 10 $\rm deg^2$., Comment: 17 pages, 11 figures, 3 tables. Accepted for publication in MNRAS

The lithium metal-solid-state electrolyte interface plays a critical role in the performance of solid-state batteries. However, operando characterization of the buried interface morphology in solid-state cells is particularly difficult because of the lack of direct optical access. Destructive techniques that require isolating the interface inadvertently modify the interface and cannot be used for operando monitoring. In this work, we introduce the concept of thermal wave sensing using modified 3-omega sensors that is attached to the outside of the lithium metal-solid state cells to non-invasively probe the morphology of the lithium metal-electrolyte interface. We show that the thermal interface resistance measured by the 3-omega sensors relates directly to the physical morphology of the interface and demonstrate that 3-omega thermal wave sensing can be used for non-invasive operando monitoring the morphology evolution of the lithium metal-solid state electrolyte interface.

The ASTRI (Astrofisica con Specchi a Tecnologia Replicante Italiana) Project led by the Italian National Institute for Astrophysics (INAF) is developing and will deploy at the Observatorio del Teide a mini-array (ASTRI Mini-Array) composed of nine telescopes similar to the small-size dual-mirror Schwarzschild-Couder telescope (ASTRI-Horn) currently operating on the slopes of Mt. Etna in Sicily. The ASTRI Mini-Array will surpass the current Cherenkov telescope array differential sensitivity above a few tera-electronvolt (TeV), extending the energy band well above hundreds of TeV. This will allow us to explore a new window of the electromagnetic spectrum, by convolving the sensitivity performance with excellent angular and energy resolution figures. In this paper we describe the Core Science that we will address during the first four years of operation, providing examples of the breakthrough results that we will obtain when dealing with current open questions, such as the acceleration of cosmic rays, cosmology and fundamental physics and the new window, for the TeV energy band, of the time-domain astrophysics., Comment: Published in Journal of High Energy Astrophysics. 46 Figures, 7 Tables

Quasars powered by supermassive black holes (MBH, $>10^8~M_{\odot}$) at $z\sim 6$ are predicted to reside in cosmic over-dense regions. However, observations so far could not confirm this expectation due to limited statistics. The picture is further complicated by the possible effects of quasar outflows (i.e. feedback) that could either suppress or stimulate the star formation rate (SFR) of companion galaxies, thus modifying the expected bias. Here we quantify feedback effects on the properties and detectability of companions by comparing cosmological zoom-in simulations of a quasar in which feedback is either included or turned-off. With respect to the no-feedback case, companions (a) directly impacted by the outflow have their SFR increased by a factor $2-3$, and (b) tend to be more massive. Both effects shift the [CII]158$\mu$m and UV luminosity functions toward brighter magnitudes. This leads us to conclude that quasar feedback slightly increases the effective quasar bias, boosting the number density of observable quasar companions, in agreement with what has been found around the brightest quasars of recent ALMA [CII] surveys. Deeper observations performed with JWST and/or ALMA will improve the statistical significance of this result by detecting a larger number of fainter quasar companions., Comment: 20 pages, 13 figures; Accepted for publications in MNRAS

Recent years have seen the adoption of Machine Learning (ML) techniques to predict the performance of large-scale applications, mostly at a coarse level. In contrast, we propose to use ML techniques for performance prediction at a much finer granularity, namely at the Basic Block (BB) level, which are single entry, single exit code blocks that are used for analysis by the compilers to break down a large code into manageable pieces. We extrapolate the basic block execution counts of GPU applications and use them for predicting the performance for large input sizes from the counts of smaller input sizes. We train a Poisson Neural Network (PNN) model using random input values as well as the lowest input values of the application to learn the relationship between inputs and basic block counts. Experimental results show that the model can accurately predict the basic block execution counts of 16 GPU benchmarks. We achieve an accuracy of 93.5% in extrapolating the basic block counts for large input sets when trained on smaller input sets and an accuracy of 97.7% in predicting basic block counts on random instances. In a case study, we apply the ML model to CUDA GPU benchmarks for performance prediction across a spectrum of applications. We use a variety of metrics for evaluation, including global memory requests and the active cycles of tensor cores, ALU, and FMA units. Results demonstrate the model's capability of predicting the performance of large datasets with an average error rate of 0.85% and 0.17% for global and shared memory requests, respectively. Additionally, to address the utilization of the main functional units in Ampere architecture GPUs, we calculate the active cycles for tensor cores, ALU, FMA, and FP64 units and achieve an average error of 2.3% and 10.66% for ALU and FMA units while the maximum observed error across all tested applications and units reaches 18.5%., Comment: Accepted at the 29th IEEE International Conference on Parallel and Distributed Systems (ICPADS 2023)

We present PPT-Multicore, an analytical model embedded in the Performance Prediction Toolkit (PPT) to predict parallel application performance running on a multicore processor. PPT-Multicore builds upon our previous work towards a multicore cache model. We extract LLVM basic block labeled memory trace using an architecture-independent LLVM-based instrumentation tool only once in an application's lifetime. The model uses the memory trace and other parameters from an instrumented sequentially executed binary. We use a probabilistic and computationally efficient reuse profile to predict the cache hit rates and runtimes of OpenMP programs' parallel sections. We model Intel's Broadwell, Haswell, and AMD's Zen2 architectures and validate our framework using different applications from PolyBench and PARSEC benchmark suites. The results show that PPT-Multicore can predict cache hit rates with an overall average error rate of 1.23% while predicting the runtime with an error rate of 9.08%., Comment: arXiv admin note: text overlap with arXiv:2103.10635. J Supercomput (2021)

Performance modeling of parallel applications on multicore processors remains a challenge in computational co-design due to multicore processors' complex design. Multicores include complex private and shared memory hierarchies. We present a Scalable Analytical Shared Memory Model (SASMM). SASMM can predict the performance of parallel applications running on a multicore. SASMM uses a probabilistic and computationally-efficient method to predict the reuse distance profiles of caches in multicores. SASMM relies on a stochastic, static basic block-level analysis of reuse profiles. The profiles are calculated from the memory traces of applications that run sequentially rather than using multi-threaded traces. The experiments show that our model can predict private L1 cache hit rates with 2.12% and shared L2 cache hit rates with about 1.50% error rate., Comment: 11 pages, 5 figures. arXiv admin note: text overlap with arXiv:1907.12666

We investigate the infrared (IR) emission of high-redshift ($z\sim 6$), highly star-forming (${ {\rm SFR} > 100}$ $M_{\rm \odot} {\rm yr}^{-1}$) galaxies, with/without Active Galactic Nuclei (AGN), using a suite of cosmological simulations featuring dust radiative transfer. Synthetic Spectral Energy Distributions (SEDs) are used to quantify the relative contribution of stars/AGN to dust heating. In dusty (${M_{\rm d}\gtrsim 3\times 10^7 M_{\rm \odot}}$) galaxies, $\gtrsim 50-90 \%$ of the UV radiation is obscured by dust inhomogeneities on scales ${\gtrsim 100}$ pc. In runs with AGN, a clumpy, warm ($\approx 250$ K) dust component co-exists with a colder ($\approx 60$ K) and more diffuse one, heated by stars. Warm dust provides up to ${50 \%}$ of the total IR luminosity, but only $\lesssim 0.1 \%$ of the total mass content. The AGN boosts the MIR flux by ${10-100 \times}$ with respect to star forming galaxies, without significantly affecting the FIR. Our simulations successfully reproduce the observed SED of bright (${M_{\rm UV}\sim -26}$) ${z\sim 6}$ quasars, and show that these objects are part of complex, dust-rich merging systems, containing multiple sources (accreting BHs and/or star forming galaxies) in agreement with recent HST and ALMA observations. Our results show that the proposed ORIGINS missions will be able to investigate the MIR properties of dusty star forming galaxies and to obtain good quality spectra of bright quasars at $z\sim 6$. Finally, the MIR-to-FIR flux ratio of faint (${M_{\rm UV}\sim -24}$) AGN is ${>10\times}$ higher than for normal star forming galaxies. This implies that combined JWST/ORIGINS/ALMA observations will be crucial to identify faint and/or dust-obscured AGN in the distant Universe., Comment: 22 pages, 13 figures, accepted for publication on MNRAS

In this paper, we shall be considering the Waring's problem for matrices. One version of the problem involves writing an $n \times n$ matrix over a commutative ring $R$ with unity as a sum of $k$-th powers of matrices over $R.$ This study is motivated by the interesting results of Carlitz, Newman, Vaserstein, Griffin, Krusemeyer, Richman etc. obtained earlier in this direction. The results are for the case $n \geq k \geq 2$ in terms of the trace of the matrix. For $n < k,$ it was shown by Katre, Garge that it is enough to work with the special case $n = 2$ and $k \geq 3.$ The cases $3 \leq k \leq 5$ and $k = 7$ were settled in earlier results. There was no case of a composite, non-prime-power $k$ occuring above. In this paper, we will find the trace criteria for a matrix to be a sum of sixth (a composite non-prime power) and eighth powers of matrices over a commutative ring $R$ with unity. An elegant discriminant criterion was obtained by Katre and Khule earlier in the special case of an order in an algebraic number field $\mathcal{O}.$ We will derive here similar discriminant criteria for every matrix over $\mathcal{O}$ to be a sum of sixth and eighth powers of matrices over $\mathcal{O}.$

Performance modeling of parallel applications on multicore computers remains a challenge in computational co-design due to the complex design of multicore processors including private and shared memory hierarchies. We present a Scalable Analytical Shared Memory Model to predict the performance of parallel applications that runs on a multicore computer and shares the same level of cache in the hierarchy. This model uses a computationally efficient, probabilistic method to predict the reuse distance profiles, where reuse distance is a hardware architecture-independent measure of the patterns of virtual memory accesses. It relies on a stochastic, static basic block-level analysis of reuse profiles measured from the memory traces of applications ran sequentially on small instances rather than using a multi-threaded trace. The results indicate that the hit-rate predictions on the shared cache are accurate.

The Cherenkov Telescope Array (CTA) is the major next-generation observatory for ground-based very-high-energy gamma-ray astronomy. It will improve the sensitivity of current ground-based instruments by a factor of five to twenty, depending on the energy, greatly improving both their angular and energy resolutions over four decades in energy (from 20 GeV to 300 TeV). This achievement will be possible by using tens of imaging Cherenkov telescopes of three successive sizes. They will be arranged into two arrays, one per hemisphere, located on the La Palma island (Spain) and in Paranal (Chile). We present here the optimised and final telescope arrays for both CTA sites, as well as their foreseen performance, resulting from the analysis of three different large-scale Monte Carlo productions., Comment: 48 pages, 16 figures, accepted for publication in Astroparticle Physics

Accretion of matter onto central Black Holes (BHs) in galaxies liberates enormous amounts of feedback energy, which influence the formation and evolution of structures, affecting the environment from pc to Mpc scales. These BHs are usually Supermassive BHs (SMBHs: mass $\geq 10^6 M_{\odot}$) existing at the centers of active galactic nuclei (AGN), which are widely observed through their multi-wavelength emission at all cosmic epochs. The SMBH energy output is often observed as powerful AGN outflows in a wide variety of forms. Relatively recently, Intermediate-Mass BHs (IMBHs: mass = $100 - 10^6 M_{\odot}$) have started to be observed hosted in Dwarf Galaxy (DG) centers. Some of the central IMBHs in DGs show signatures of activity in the form of low-luminosity AGN. We have performed Cosmological Hydrodynamical Simulations to probe SMBHs in high-z quasars (Barai et al. 2018), and IMBHs in DGs (Barai & de Gouveia Dal Pino 2019). Our simulations employ the 3D TreePM SPH code GADGET-3, and include metal cooling, star formation, chemical enrichment, stellar evolution, supernova feedback, AGN accretion and feedback. Analyzing the simulation output in post-processing, we investigate the growth of the first IMBHs, and the growth of the first SMBHs, their impact on star-formation, as well as their co-evolution with the respective host galaxies. We quantify the impact of SMBHs and IMBHs on their host galaxies, especially the effects on quenching star-formation. We also study the corresponding BH outflow properties., Comment: 11 pages, Conference Proceedings of Talk presented at the BHCB 2018 - "International Conference on Black Holes as Cosmic Batteries: UHECRs and Multimessenger Astronomy", held at Foz do Igua\c{c}u (Brazil), 2018 September 12-15