arXiv subjects

Arun Kumar

Publications and source records attributed to Arun Kumar.

At least 55 records · Page 3Linked to original sources

arXiv2023

An Optimized Tri-store System for Multi-model Data Analytics

Data science applications increasingly rely on heterogeneous data sources and analytics. This has led to growing interest in polystore systems, especially analytical polystores. In this work, we focus on a class of emerging multi-data model analytics workloads that fluidly straddle relational, graph, and text analytics. Instead of a generic polystore, we build a ``tri-store'' system that is more aware of the underlying data models to better optimize execution to improve scalability and runtime efficiency. We name our system AWESOME (Analytics WorkbEnch for SOcial MEdia). It features a powerful domain-specific language named ADIL. ADIL builds on top of underlying query engines (e.g., SQL and Cypher) and features native data types for succinctly specifying cross-engine queries and NLP operations, as well as automatic in-memory and query optimizations. Using real-world tri-model analytical workloads and datasets, we empirically demonstrate the functionalities of AWESOME for scalable data science applications and evaluate its efficiency.

cs.DB↗

arXiv2023

Humbert Generalized Fractional Differenced ARMA Processes

In this article, we use the generating functions of the Humbert polynomials to define two types of Humbert generalized fractional differenced ARMA processes. We present stationarity and invertibility conditions for the introduced models. The singularities for the spectral densities of the introduced models are obtained. In particular, Pincherle ARMA, Horadam ARMA and Horadam-Pethe ARMA processes are studied.

math.ST↗

arXiv2023

Nonsingular black hole chemistry in $4D$ Einstein-Gauss-Bonnet gravity

The EGB is an outcome of quadratic curvature corrections to the Einstein-Hilbert gravity action in the form of a Gauss-Bonnet (GB) term in $ D > 4$ dimensions, and EGB gravity is topologically invariant in $4D$. Several ways have been proposed for regularizing the $ D \to 4 $ limit of EGB for non-trivial gravitational dynamics in $ 4D $. Motivated by the importance of AdS/CFT, we obtain an exact static spherically symmetric nonsingular black hole in $4D$ EGB gravity coupled to the nonlinear electrodynamics (NED) in an AdS spacetime. We interpret the negative cosmological constant $Λ$ as the positive pressure, via $ P=-Λ/8π$, of the system's thermodynamic properties of the nonsingular black hole with an AdS background. We find that for $P 0$ are stable to thermal fluctuations and unstable otherwise. We also analyzed the Gibbs free energy to find that the small globally unstable black holes undergo a phase transition to the large globally stable black holes. Further, we study the $P-V$ criticality of the system and then calculate the critical exponents to find that our system behaves like Van der Walls fluid.

gr-qc↗

arXiv2022

Objects as Spatio-Temporal 2.5D points

Determining accurate bird's eye view (BEV) positions of objects and tracks in a scene is vital for various perception tasks including object interactions mapping, scenario extraction etc., however, the level of supervision required to accomplish that is extremely challenging to procure. We propose a light-weight, weakly supervised method to estimate 3D position of objects by jointly learning to regress the 2D object detections and scene's depth prediction in a single feed-forward pass of a network. Our proposed method extends a center-point based single-shot object detector, and introduces a novel object representation where each object is modeled as a BEV point spatio-temporally, without the need of any 3D or BEV annotations for training and LiDAR data at query time. The approach leverages readily available 2D object supervision along with LiDAR point clouds (used only during training) to jointly train a single network, that learns to predict 2D object detection alongside the whole scene's depth, to spatio-temporally model object tracks as points in BEV. The proposed method is computationally over $\sim$10x efficient compared to recent SOTA approaches while achieving comparable accuracies on KITTI tracking benchmark.

cs.CV↗

arXiv2022

Technology Pipeline for Large Scale Cross-Lingual Dubbing of Lecture Videos into Multiple Indian Languages

Cross-lingual dubbing of lecture videos requires the transcription of the original audio, correction and removal of disfluencies, domain term discovery, text-to-text translation into the target language, chunking of text using target language rhythm, text-to-speech synthesis followed by isochronous lipsyncing to the original video. This task becomes challenging when the source and target languages belong to different language families, resulting in differences in generated audio duration. This is further compounded by the original speaker's rhythm, especially for extempore speech. This paper describes the challenges in regenerating English lecture videos in Indian languages semi-automatically. A prototype is developed for dubbing lectures into 9 Indian languages. A mean-opinion-score (MOS) is obtained for two languages, Hindi and Tamil, on two different courses. The output video is compared with the original video in terms of MOS (1-5) and lip synchronisation with scores of 4.09 and 3.74, respectively. The human effort also reduces by 75%.

eess.AS↗

arXiv2022

GARTFIMA Process and its Empirical Spectral Density Based Estimation

In this article, we introduce a Gegenbauer autoregressive tempered fractionally integrated moving average (GARTFIMA) process. We work on the spectral density and autocovariance function for the introduced process. The parameter estimation is done using the empirical spectral density with the help of the nonlinear least square technique and the Whittle likelihood estimation technique. The performance of the proposed estimation techniques is assessed on simulated data. Further, the introduced process is shown to better model the real-world data in comparison to other time series models.

math.ST↗

arXiv2022

Hydra: A System for Large Multi-Model Deep Learning

Scaling up model depth and size is now a common approach to raise accuracy in many deep learning (DL) applications, as evidenced by the widespread success of multi-billion or even trillion parameter models in natural language processing (NLP) research. Despite success in DL research and at major technology companies, broader practical adoption of such large models among domain scientists and businesses is still bottlenecked by GPU memory limits, high training costs, and low GPU availability, even on public clouds. Model selection needs further compound these resource challenges: users often need to compare dozens of models with different hyper-parameters or neural architectures to suit their specific task and dataset. In this paper, we present Hydra, a system designed to tackle such challenges by enabling out-of-the-box scaling for multi-large-model DL workloads on even commodity GPUs in a resource-efficient manner. Hydra is the first approach to holistically optimize the execution of multi-model workloads for large DL models. We do this by adapting prior "model-parallel" execution schemes to work with scalable parameter offloading across the memory hierarchy and further hybridizing this approach with task-parallel job scheduling techniques. Hydra decouples scalability of model parameters from parallelism of execution, thus enabling DL users to train even a 6-billion parameter model on a single commodity GPU. It also fully exploits the speedup potential of task parallelism in multi-GPU setups, yielding near-linear strong scaling and making rigorous model selection perhaps more practical for such models. We evaluate end-to-end performance by fine-tuning GPT-2 for language modeling. We find that Hydra offers between 50% and 100% higher training throughput than even the best settings of state-of-the-art industrial frameworks such as DeepSpeed and GPipe for multi-large-model training.

cs.DC↗

arXiv2022

AWESOME: Empowering Scalable Data Science on Social Media Data with an Optimized Tri-Store Data System

Modern data science applications increasingly use heterogeneous data sources and analytics. This has led to growing interest in polystore systems, especially analytical polystores. In this work, we focus on emerging multi-data model analytics workloads over social media data that fluidly straddle relational, graph, and text analytics. Instead of a generic polystore, we build a "tri-store" system that is more aware of the underlying data models to better optimize execution to improve scalability and runtime efficiency. We name our system AWESOME (Analytics WorkbEnch for SOcial MEdia). It features a powerful domain-specific language named ADIL. ADIL builds on top of underlying query engines (e.g., SQL and Cypher) and features native data types for succinctly specifying cross-engine queries and NLP operations, as well as automatic in-memory and query optimizations. Using real-world tri-model analytical workloads and datasets, we empirically demonstrate the functionalities of AWESOME for scalable data science over social media data and evaluate its efficiency.

cs.DB↗

arXiv2022

Bardeen Black Holes in the Regularized $4D$ Einstein--Gauss--Bonnet Gravity

We obtain exact Bardeen black holes to the regularized $4D$ Einstein--Gauss--Bonnet (EGB) gravity minimally coupled with the nonlinear electrodynamics (NED). In turn, we analyze the horizon structure to determine the effect of GB parameter $α$ on the minimum cutoff values of mass, $M_0$, and magnetic monopole charge, $g_0$, for the existence of a black hole horizon. We obtain an exact expression for thermodynamic quantities, namely, Hawking temperature $T_+$, entropy $S_+$, Helmholtz free energy $F_+$, and specific heat $C_+$ associated with the black hole horizon, and they show significant deviations from the $4D$ EGB case owing to NED. Interestingly, there exists a critical value of horizon radius, $r_+^{c}$, corresponding to the local maximum of Hawking temperature, at which heat capacity diverges, confirming the second-order phase transition. A discussion on the black holes of alternate regularized $4D$ EGB gravity belonging to the scalar-tensor theory is appended.

gr-qc↗

arXiv2022

Nonsingular Black Holes in $4D$ Einstein--Gauss--Bonnet Gravity

Recently, several methods have been proposed to regularize a $D \to 4$ limit of Einstein-Gauss-Bonnet (EGB), leading to nontrivial gravitational dynamics in $4D$. We present an exact nonsingular black hole solution in the $4D$ EGB gravity coupled to non-linear electrodynamics and analyze their thermodynamic properties to calculate precise expressions for the black hole mass, temperature, and entropy. Because of the magnetic charge, the thermodynamic quantities are corrected, and the Hawking--Page phase transition is achievable with diverges of the heat capacity at a larger critical radius $r=r_{+}^{C}$ in comparison to the $5D$ counterpart where the temperature is maximum. Thus, we have a black hole with Cauchy and event horizons, and its evaporation leads to a thermodynamically stable extremal black hole remnant with vanishing temperature, and its size is larger than the $5D$ counterpart. The entropy does not satisfy the usual exact horizon Bekenstein--Hawking area law of general relativity with a logarithmic area correction term.

gr-qc↗

arXiv2022

Thermodynamic stability and P-V criticality of nonsingular-AdS black holes endowed with clouds of strings

We investigate the extended phase space thermodynamics of nonsingular-AdS black holes minimally coupled to clouds of strings in which we consider the cosmological constant ($Λ$) as the pressure ($P$) of the black holes and its conjugate variable thermodynamical volume ($V$) of the black holes. Owing to the background clouds of strings parameter ($a$), we analyse the Hawking temperature, entropy and specific heat on horizon radius for fixed-parameter $k$. We find that the strings clouds background does not alter small/large black hole (SBH/LBH) phase transition but occurs at a larger horizon radius, and two second-order phase transitions occur at a smaller horizon radius. Indeed, the $G$--$T$ plots exhibit a swallowtail below the critical pressure, implying that the first-order phase transition is analogous to the liquid-gas phase transition at a lower temperature and lower critical pressure. To further examine the analogy between nonsingular-AdS black holes and a liquid-gas system, we derive the exact critical points and probe the effects of a cloud of strings on $P-V$ criticality to find that the isotherms undergo liquid-gas like phase transition for $\tilde{T}\,<\,\tilde{T}_c$ at lower $\tilde{T}_c$. We have also calculated the critical exponents identical with Van der Walls fluid, i.e., same as those obtained before for arbitrary other AdS black holes, which implies that the background clouds of strings do not change the critical exponents.

gr-qc↗

arXiv2022

Figure-of-merit for Semi-transparent Solar Cells

Semi-transparent Solar Cells (ST-SCs) has emerged as one of the most prominent energy harvesting technology that combines the benefits of light transparency and light-to-electricity conversion. The biggest opportunities for such technologies lie in their integration as windows and skylights within energy-sustainable buildings or combining them with other solar cell technologies in tandem configuration. The performance of ST-SCs is mainly determined by the trade-off between the competing parameters of the capability to convert the incident light into electricity while allowing some parts to transmit providing transparency through the device. Depending on the target application, the selection of ST-SCs is a tricky affair as some devices might offer high efficiency but compromises transparency and vice-versa. On the other way around, this is again not helped by the fact that due to advancements in materials engineering, processing, and characterization, a vastly different combination of efficiency and transparency has been reported by research groups. So, in order to quantify the performance of ST-SCs, we proposed, a figure-of-merit (FoM) which can be used as a tool that can help in analysing and comparing the performance among various ST-SCs. The defined FoM focuses on the power conversion efficiency of the device, bifaciality factor, transmittance in the desired region, and that corresponding to 550 nm wavelength. Additionally, in this work, we have been shown how the proposed FoM can be correlated for tandem and building-integrated photovoltaics applications. Based on these resultant parameters, FoM is calculated and compared for different device architectures available in the literature. The proposed FoM shall serve as a meaningful guiding path to the researchers for the development of advanced ST-SCs.

cond-mat.mtrl-sci↗

arXiv2021

The expectation-maximization algorithm for autoregressive models with normal inverse Gaussian innovations

The autoregressive (AR) models are used to represent the time-varying random process in which output depends linearly on previous terms and a stochastic term (the innovation). In the classical version, the AR models are based on normal distribution. However, this distribution does not allow describing data with outliers and asymmetric behavior. In this paper, we study the AR models with normal inverse Gaussian (NIG) innovations. The NIG distribution belongs to the class of semi heavy-tailed distributions with wide range of shapes and thus allows for describing real-life data with possible jumps. The expectation-maximization (EM) algorithm is used to estimate the parameters of the considered model. The efficacy of the estimation procedure is shown on the simulated data. A comparative study is presented, where the classical estimation algorithms are also incorporated, namely, Yule-Walker and conditional least squares methods along with EM method for model parameters estimation. The applications of the introduced model are demonstrated on the real-life financial data.

stat.ME↗

arXiv2021

Densities of Inverse Tempered Stable Subordinators and Related Processes With Mellin Transforrm

In this article, the infinite series form of the probability densities of tempered stable and inverse tempered stable subordinators are obtained using Mellin transform. Further, the densities of the products and quotients of stable and inverse stable subordinators are worked out. The asymptotic behaviours of these densities are obtained as $x \rightarrow 0^+$. Similar results for tempered and inverse tempered stable subordinators are discussed. Our results provide alternative methods to find the densities of these subordinators and complement the results available in literature.

math.PR↗

arXiv2021

Tempered Stable Autoregressive Models

In this article, we introduce and study a one sided tempered stable first order autoregressive model called TAR(1). Under the assumption of stationarity of the model, the marginal probability density function of the error term is found. It is shown that the distribution of the error term is infinitely divisible. Parameter estimation of the introduced TAR(1) process is done by adopting the conditional least square and method of moments based approach and the performance of the proposed methods are evaluated on simulated data. Also we study an autoregressive model of order one with tempered stable innovations. Using appropriate test statistic it is shown that the model fit very well on real and simulated data. Our models generalize the inverse Gaussian and one-sided stable autoregressive models existing in the literature.

math.ST↗

arXiv2021

Normal Inverse Gaussian Autoregressive Model Using EM Algorithm

In this article, normal inverse Gaussian (NIG) autoregressive model is introduced. The parameters of the model are estimated using Expectation Maximization (EM) algorithm. The efficacy of the EM algorithm is shown using simulated and real world financial data. It is shown that NIG autoregressive model fit very well the considered financial data and hence could be useful in modeling of various real life time-series data.

stat.ME↗

arXiv2021

Nonsingular black hole chemistry

We study the nonsingular black hole in Anti de-Sitter background taking the negative cosmological constant as the pressure of the system. We investigate the horizon structure, and find the critical values $m_0$ and $\tilde{k}_0$, such that $m>m_0$ (or $\tilde{k}<\tilde{k}_0$) corresponds to a black solution with two horizons, namely the Cauchy horizon $x_-$ and the event horizon $x_+$. For $m=m_0$ (or $\tilde{k}=\tilde{k}_0$), there exist an extremal black hole with degenerate horizon $x_0=x_{\pm}$ and for $m \tilde{k}_0$), no black hole solution exists. In turn, we calculate the thermodynamical properties and by observing the behaviour of Gibb's free energy and specific heat, we find that this black hole solution exhibits first order (small to large black hole) and second order phase transition. Further, we study the $P-V$ criticality of system and then calculate the critical exponents showing that they are the same as those of the Van der Waals fluid.

gr-qc↗

arXiv2021

Fractional Poisson Processes of Order k and Beyond

In this article, we introduce fractional Poisson felds of order k in n-dimensional Euclidean space $R_n^+$. We also work on time-fractional Poisson process of order k, space-fractional Poisson process of order k and tempered version of time-space fractional Poisson process of order k in one dimensional Euclidean space $R_1^+$. These processes are defined in terms of fractional compound Poisson processes. Time-fractional Poisson process of order k naturally generalizes the Poisson process and Poisson process of order k to a heavy tailed waiting times counting process. The space-fractional Poisson process of order k, allows on average infinite number of arrivals in any interval. We derive the marginal probabilities, governing difference-differential equations of the introduced processes. We also provide Watanabe martingale characterization for some time-changed Poisson processes.

math.PR↗