arXiv subjects

Supriyo Ghosh

Publications and source records attributed to Supriyo Ghosh.

At least 55 records · Page 3Linked to original sources

arXiv2022

Neural-Progressive Hedging: Enforcing Constraints in Reinforcement Learning with Stochastic Programming

We propose a framework, called neural-progressive hedging (NP), that leverages stochastic programming during the online phase of executing a reinforcement learning (RL) policy. The goal is to ensure feasibility with respect to constraints and risk-based objectives such as conditional value-at-risk (CVaR) during the execution of the policy, using probabilistic models of the state transitions to guide policy adjustments. The framework is particularly amenable to the class of sequential resource allocation problems since feasibility with respect to typical resource constraints cannot be enforced in a scalable manner. The NP framework provides an alternative that adds modest overhead during the online phase. Experimental results demonstrate the efficacy of the NP framework on two continuous real-world tasks: (i) the portfolio optimization problem with liquidity constraints for financial planning, characterized by non-stationary state distributions; and (ii) the dynamic repositioning problem in bike sharing systems, that embodies the class of supply-demand matching problems. We show that the NP framework produces policies that are better than deep RL and other baseline approaches, adapting to non-stationarity, whilst satisfying structural constraints and accommodating risk measures in the resulting policies. Additional benefits of the NP framework are ease of implementation and better explainability of the policies.

cs.LG↗

arXiv2022

$H$-band Temperature and Metallicity Indicators for Cool Giants: Empirical Relations in Bayesian Framework

We explored here the near-infrared $H$-band atmospheric window aiming to provide quantitative diagnostic tools for deriving stellar parameters, for instance, effective temperature ($T_{eff}$) and metallicity ([$Fe/H$]), of cool giants ($T_{eff}$ $<$ 5000 K) using low-resolution spectra. We obtained 177 cool giants from the X-shooter spectral library covering a wider metallicity range ($-$2.35 dex $<$ [$Fe/H$] $<$ 0.5 dex) than in earlier works. Degrading the spectral resolution to R$\sim$ 1200, we estimated equivalent widths of several important spectral features, and the behavior of spectral features with stellar parameters are studied. Also, the empirical relations for deriving $T_{eff}$ and [$Fe/H$] are established in the Bayesian framework. We found that $^{12}$CO at 1.56 $μ$m and 1.62 $μ$m, and $^{12}$CO+MgI at 1.71 $μ$m are the best three $T_{eff}$ indicators with a typical accuracy of 153 K, 123 K and 107 K, respectively. The cubic Bayesian model provides the best metallicity estimator with a typical accuracy of 0.22 dex, 0.28 dex, and 0.24 dex for FeH at 1.62 $μ$m, $^{12}$CO at 1.64 $μ$m, and Fe I at 1.66 $μ$m, respectively. We also showed a detailed quantitative metallicity dependence of $T_{eff}-$EWs correlations defining three metallicity groups, supersolar ([$Fe/H$] $>$ 0.0 dex), solar ($-$0.3 dex $<$ [$Fe/H$] $<$ 0.3 dex), and subsolar ([$Fe/H$] $<$ $-$0.3 dex), from Hierarchical Bayesian modelling. The difference between the solar and subsolar relationship is statistically significant, but such difference is not evident between the solar and supersolar groups.

astro-ph.SR↗

arXiv2021

TIFR Treasures for Astronomy from Ground to Space

The infrared astronomy group of Department of Astronomy and Astrophysics at Tata Institute of Fundamental Research has been pursuing astronomical instrumentation activities since its inception. The group has been routinely involved in balloon-borne astronomy programs from the field station at Hyderabad with indigenously developed payloads. Ground-based astronomical activities began with a single element infrared detector. Later, over time, larger format array detectors are being used in the cameras. These astronomy cameras have been routinely used at observatories across India. Recently, the group has also developed a laboratory model of the Infrared Spectroscopic Imaging Survey payload, targeted for the small satellite mission of the Indian Space Research Organisation, which will carry out spectroscopic measurements in the wavelength range 1.7 to 6.4 $μ$m seamlessly

astro-ph.IM↗

arXiv2021

Tusas: A fully implicit parallel approach for coupled phase-field equations

We develop a fully-coupled, fully-implicit approach for phase-field modeling of solidification in metals and alloys. Predictive simulation of solidification in pure metals and metal alloys remains a significant challenge in the field of materials science, as microstructure formation during the solidification process plays a critical role in the properties and performance of the solid material. Our simulation approach consists of a finite element spatial discretization of the fully-coupled nonlinear system of partial differential equations at the microscale, which is treated implicitly in time with a preconditioned Jacobian-free Newton-Krylov method. The approach allows time steps larger than those restricted by the traditional explicit CFL limit and is algorithmically scalable as well as efficient due to an effective preconditioning strategy based on algebraic multigrid and block factorization. We implement this approach in the open-source Tusas framework, which is a general, flexible tool developed in C++ for solving coupled systems of nonlinear partial differential equations. The performance of our approach is analyzed in terms of algorithmic scalability and efficiency, while the computational performance of Tusas is presented in terms of parallel scalability and efficiency on emerging heterogeneous architectures. We demonstrate that modern algorithms, discretizations, and computational science, and heterogeneous hardware provide a robust route for predictive phase-field simulation of microstructure evolution during additive manufacturing.

cs.CE↗

arXiv2021

Picking Pearl From Seabed: Extracting Artefacts from Noisy Issue Triaging Collaborative Conversations for Hybrid Cloud Services

Site Reliability Engineers (SREs) play a key role in issue identification and resolution. After an issue is reported, SREs come together in a virtual room (collaboration platform) to triage the issue. While doing so, they leave behind a wealth of information which can be used later for triaging similar issues. However, usability of the conversations offer challenges due to them being i) noisy and ii) unlabelled. This paper presents a novel approach for issue artefact extraction from the noisy conversations with minimal labelled data. We propose a combination of unsupervised and supervised model with minimum human intervention that leverages domain knowledge to predict artefacts for a small amount of conversation data and use that for fine-tuning an already pretrained language model for artefact prediction on a large amount of conversation data. Experimental results on our dataset show that the proposed ensemble of unsupervised and supervised model is better than using either one of them individually.

cs.AI↗

arXiv2021

Spectroscopic and photometric monitoring of a poorly known high-luminous OH/IR star: IRAS 18278+0931

We present the time-dependent properties of a poorly known OH/IR star $-$ IRAS 18278+0931 (hereafter, IRAS 18+09) towards the Ophiuchus constellation. We have carried out long-term optical/near-infrared (NIR) photometric and spectroscopic observations to study the object. From optical $R$- and $I$-band light curves, the period of IRAS 18+09 is estimated to be 575 $\pm$ 30 days and the variability amplitudes range from $Δ$R $\sim$ 4.0 mag to $Δ$I $\sim$ 3.5 mag. From the standard Period-Luminosity (PL) relations, the distance ($D$) to the object, 4.0 $\pm$ 1.3 kpc, is estimated. Applying this distance in the radiative transfer model, the spectral energy distribution (SED) are constructed from multi-wavelength photometric and IRAS-LRS spectral data which provides the luminosity, optical depth, and gas mass-loss rate (MLR) of the object to be 9600 $\pm$ 500 $L_{\odot}$, 9.1 $\pm$ 0.6 at 0.55 $μ$m and 1.0$\times$10$^{-6}$ M$_\odot$ yr$^{-1}$, respectively. The current mass of the object infers in the range 1.0 $-$ 1.5 $M_\odot$ assuming solar metallicity. Notably, the temporal variation of atomic and molecular features (e.g., TiO, Na I, Ca I, CO, H$_2$O) over the pulsation cycle of the OH/IR star illustrates the sensitivity of the spectral features to the dynamical atmosphere as observed in pulsating AGB stars.

astro-ph.SR↗

arXiv2020

New Temperature and Metallicity Scale of Cool Giants from K-band Spectra

We present here quantitative diagnostic tools for cool giants that employ low-resolution near-infrared spectroscopy in the $K$-band for stellar population studies. In this study, a total of 260 cool giants (177 stars observed with X-shooter and 83 stars observed with NIFS) are used covering a wider metallicity range than in earlier works. We measure equivalent widths of some of the selected important $K$-band spectral features like Na I, Fe I, and $^{12}$CO after degrading the spectral resolution (R $\sim$ 1200) to investigate the spectral behavior with fundamental parameters (e.g. effective temperature and metallicity). We derive empirical relations to measure effective temperature using the $^{12}$CO first-overtone band at 2.29 $μ$m and 2.32 $μ$m and show a detailed quantitative metallicity dependence of these correlations. We find that the empirical relations based on solar-neighborhood stars can incorporate large uncertainty in evaluating $T_{eff}$ for metal-poor or metal-rich stars. Furthermore, we explore all the spectral lines to establish the empirical relation with metallicity and find that the quadratic fit of the combination of Na I and $^{12}$CO at 2.29 $μ$m lines yields a reliable empirical relation at [$Fe/H$] $\leq$ $-$0.4 dex, while a linear fit of any line offers a good metallicity scale for stars having [$Fe/H$] $\geq$ 0.0 dex.

astro-ph.SR↗

arXiv2020

An Iterative Security Game for Computing Robust and Adaptive Network Flows

The recent advancement in real-world critical infrastructure networks has led to an exponential growth in the use of automated devices which in turn has created new security challenges. In this paper, we study the robust and adaptive maximum flow problem in an uncertain environment where the network parameters (e.g., capacities) are known and deterministic, but the network structure (e.g., edges) is vulnerable to adversarial attacks or failures. We propose a robust and sustainable network flow model to effectively and proactively counter plausible attacking behaviors of an adversary operating under a budget constraint. Specifically, we introduce a novel scenario generation approach based on an iterative two-player game between a defender and an adversary. We assume that the adversary always takes a best myopic response (out of some feasible attacks) against the current flow scenario prepared by the defender. On the other hand, we assume that the defender considers all the attacking behaviors revealed by the adversary in previous iterations in order to generate a new conservative flow strategy that is robust (maximin) against all those attacks. This iterative game continues until the objectives of the adversary and the administrator both converge. We show that the robust network flow problem to be solved by the defender is NP-hard and that the complexity of the adversary's decision problem grows exponentially with the network size and the adversary's budget value. We propose two principled heuristic approaches for solving the adversary's problem at the scale of a large urban network. Extensive computational results on multiple synthetic and real-world data sets demonstrate that the solution provided by the defender's problem significantly increases the amount of flow pushed through the network and reduces the expected lost flow over four state-of-the-art benchmark approaches.

cs.GT↗

arXiv2020

Impact of Particle Arrays on Phase Separation Composition Patterns

We examine the symmetry-breaking effect of fixed constellations of particles on the surface-directed spinodal decomposition of binary blends in the presence of particles whose surfaces have a preferential affinity for one of the components. Our phase-field simulations indicate that the phase separation morphology in the presence of particle arrays can be tuned to have a continuous, droplet, lamellar, or hybrid morphology depending on the interparticle spacing, blend composition, and time. In particular, when the interparticle spacing is large compared to the spinodal wavelength, a transient target pattern composed of alternate rings of preferred and non-preferred phases emerge at early times, tending to adopt the symmetry of the particle configuration. We reveal that such target patterns stabilize for certain characteristic length, time, and composition scales characteristic of the pure phase separating mixture. To illustrate the general range of phenomena exhibited by mixture-particle systems, we simulate the effects of single-particle, multi-particle, and cluster-particle systems having multiple geometrical configurations of the particle characteristic of pattern substrates on phase separation. Our simulations show that tailoring the particle configuration, or substrate pattern configuration, a relative fluid-particle composition should allow the desirable control of the phase separation morphology as in block copolymer materials, but where the scales accessible to this approach of organizing phase-separated fluids usually are significantly larger. Limited experiments confirm the trends observed in our simulations, which should provide some guidance in engineering patterned blend and other mixtures of technological interest.

physics.chem-ph↗

arXiv2020

Effects of particles on spinodal decomposition: A Phase field study

In this thesis, we study the interplay of phase separation and wetting in multicomponent systems. For this purpose, we have examined the phase separation pattern of a binary mixture (AB) in presence of stationary spherical particles (C) which prefers one of the components of the binary (say, A). Binary AB is composed of critical composition(50:50) and off-critical compositions(60:40, 40:60). Particle sizes of 8 units and 16 units are used in the simulations. Two types of particle loading are used, 5\% and 10\%. We have employed a ternary form of Cahn-Hilliard equation to incorporate immobile fillers in our system. To elucidate the effect of wetting on phase separation we have designed three sets of $χ_{ij}$ and $κ_{ij}$ to include the effects of neutral preference, weak preference and strong preference of the particle for one of the binary components. If the particles are preferentially wetted by one of the components then early stage microstructures show transient concentric alternate layers of preferred and non-preferred phases around the particles. When particles are neutral to binary components then such a ring pattern does not form. At late times, neutral preference between particles and binary components yields a continuous morphology whereas preferential wetting produces isolated domains of non-preferred phases dispersed in a continuous matrix of preferred phase. For off-critical compositions, if minor component wets the particle then a bicontinuous morphology results whereas if major component wets the network a droplet morphology is seen. When majority component wets the particle, a possibility of double phase separation is reported. In such alloys phase separation starts near the particle surface and propagates to the bulk at intermediate to late times forming spherical or nearly spherical droplets of the minor component.

cond-mat.soft↗

arXiv2020

A Deep Ensemble Multi-Agent Reinforcement Learning Approach for Air Traffic Control

Air traffic control is an example of a highly challenging operational problem that is readily amenable to human expertise augmentation via decision support technologies. In this paper, we propose a new intelligent decision making framework that leverages multi-agent reinforcement learning (MARL) to dynamically suggest adjustments of aircraft speeds in real-time. The goal of the system is to enhance the ability of an air traffic controller to provide effective guidance to aircraft to avoid air traffic congestion, near-miss situations, and to improve arrival timeliness. We develop a novel deep ensemble MARL method that can concisely capture the complexity of the air traffic control problem by learning to efficiently arbitrate between the decisions of a local kernel-based RL model and a wider-reaching deep MARL model. The proposed method is trained and evaluated on an open-source air traffic management simulator developed by Eurocontrol. Extensive empirical results on a real-world dataset including thousands of aircraft demonstrate the feasibility of using multi-agent RL for the problem of en-route air traffic control and show that our proposed deep ensemble MARL method significantly outperforms three state-of-the-art benchmark approaches.

cs.LG↗

arXiv2020

Finite Interface Dissipation Phase Field Modeling of Ni-Nb under Additive Manufacturing Conditions

During the laser powder bed fusion (L-PBF) process, the built part undergoes multiple rapid heating-cooling cycles, leading to complex microstructures with nonuniform properties. In the present work, a computational framework, which weakly couples a finite element thermal model to a non-equilibrium PF model was developed to investigate the rapid solidification microstructure of a Ni-Nb alloy during L-PBF. The framework is utilized to predict the spatial variation of the morphology and size of cellular segregation structure as well as the microsegregation in single-track melt pool microstructures obtained under different process conditions. A solidification map demonstrating the variation of microstructural features as a function of the temperature gradient and growth rate is presented. A planar to cellular transition is predicted in the majority of keyhole mode melt pools, while a planar interface is predominant in conduction mode melt pools. The predicted morphology and size of the cellular segregation structure agrees well with experimental measurements.

physics.app-ph↗

arXiv2020

Understanding Physical Properties of Young M-dwarfs: NIR spectroscopic studies

We present here medium resolution (λ/Δλ\sim1200) H- and K-band spectra of M type dwarf stars covering the wavelength range 1.50 - 1.80 μm and 1.95 - 2.45 μm. The sample includes 53 dwarf stars (M0V-M7V) from new observations using the TIFR Near-Infrared Spectrometer and Imager (TIRSPEC) instrument on the 2-m Himalayan Chandra Telescope (HCT). Using interferometrically-measured effective temperature (T_{eff}), radius and luminosity of nearby bright calibrator stars, we have created new empirical relationships among those fundamental parameters and spectral indices. The equivalent widths of H-band spectral features like Mg (1.57 μm), Al (1.67 μm) and Mg (1.71 μm), and the H_{2}O-H index are found to be good indicators of T_{eff}, radius and luminosity and we establish the linear functions using these features relating to those stellar parameters. The root mean squared error (RMSE) of our best fits are 102K, 0.027R_{\sun} and 0.12dex respectively. Using spectral type standards along with known parallaxes, we calibrate both H and K-band H_{2}O indices as a tracer of spectral type and absolute K_{s} magnitude. Metallicities of M-dwarf samples are estimated using the K-band calibration relationships. The mass of M-dwarfs could be determined using the luminosity (L/L_{\sun}) and we establish a new empirical relation for this. We also compare and contrast our results with other similar work from the literature.

astro-ph.SR↗

arXiv2019

Exact relaxation dynamics and quantum information scrambling in multiply quenched harmonic chains

The quantum dynamics of isolated systems under quench condition exhibits a variety of interesting features depending on the integrable/chaotic nature of system. We study the exact dynamics of trivially integrable system of harmonic chains under a multiple quench protocol. Out of time ordered correlator of two Hermitian operators at large time displays scrambling in the thermodynamic limit. In this limit, the entanglement entropy and the central component of momentum distribution both saturate to a steady state value. We also show that reduced density matrix assumes the diagonal form long after multiple quenches for large system size. These exact results involving infinite dimensional Hilbert space are indicative of local thermal behaviour for a trivially integrable harmonic chain.

quant-ph↗

arXiv2019

Spectral Calibration of K$-$M Giants from medium resolution near-infrared HK-band spectra

We present here new medium resolution spectra ($λ/Δλ$ $\sim$ 1200) of K$-$M giants covering wavelength range 1.50$-$1.80 and 1.95$-$2.45 $μ$m. The sample includes 72 K0$-$M8 giants from our TIRSPEC observations and all available 35 giants in that spectral range from archival IRTF spectral library. We have calibrated here the empirical relations between fundamental parameters (e.g., effective temperature, surface gravity) and equivalent widths of some important spectral features like Si I, Na I, Ca I, $^{12}$CO. We find that the $^{12}$CO first overtone band at 2.29 $μ$m and second overtone band at 1.62 $μ$m are a reasonably good indicator of temperature above 3400 K and surface gravity. We show that the dispersion of empirical relations between $^{12}$CO and $T_{eff}$ significantly improve considering the effect of surface gravity.

astro-ph.SR↗

arXiv2019

Uncertainty Analysis of Microsegregation during Laser Powder Bed Fusion

Quality control in additive manufacturing can be achieved through variation control of the quantity of interest (QoI). We choose in this work the microstructural microsegregation to be our QoI. Microsegregation results from the spatial redistribution of a solute element across the solid-liquid interface that forms during solidification of an alloy melt pool during the laser powder bed fusion process. Since the process as well as the alloy parameters contribute to the statistical variation in microstructural features, uncertainty analysis of the QoI is essential. High-throughput phase-field simulations estimate the solid-liquid interfaces that grow for the melt pool solidification conditions that were estimated from finite element simulations. Microsegregation was determined from the simulated interfaces for different process and alloy parameters. Correlation, regression, and surrogate model analyses were used to quantify the contribution of different sources of uncertainty to the QoI variability. We found negligible contributions of thermal gradient and Gibbs-Thomson coefficient and considerable contributions of solidification velocity, liquid diffusivity, and segregation coefficient on the QoI. Cumulative distribution functions and probability density functions were used to analyze the distribution of the QoI during solidification. Our approach, for the first time, identifies the uncertainty sources and frequency densities of the QoI in the solidification regime relevant to additive manufacturing.

physics.app-ph↗

arXiv2019

Influence of morphological instability on grain boundary trajectory during directional solidification

The interplay between the diffusion-controlled dynamics of a solidification front and the trajectory of a grain boundary groove at the solid-liquid interface is studied by means of thin-sample directional solidification experiments of a transparent alloy, and by numerical simulations with the phase-field method in two dimensions. We find that low-angle grain boundaries (subboundaries) with an anisotropic interfacial free energy grow tilted at an angle $θ_t$ with respect to the temperature gradient axis. $θ_t$ remains essentially equal to its value imposed at equilibrium as long as the solidification velocity $V$ remains low. When $V$ increases and approaches the cellular instability threshold, $θ_t$ decreases, and eventually vanishes when a steady-state cellular morphology forms. The absence of mobility of the subboundary in the solid is key to this transition. These findings are in good agreement with a recent linear-stability analysis of the problem.

cond-mat.mtrl-sci↗

arXiv2018

On the Interfacial Phase Growth and Vacancy Evolution during Accelerated Electromigration in Cu/Sn/Cu Microjoints

In this work, we integrate different computational tools based on multi-phase-field simulations to account for the evolution of morphologies and crystallographic defects of Cu/Sn/Cu sandwich interconnect structures that are widely used in three dimensional integrated circuits (3DICs). Specifically, this work accounts for diffusion-driven formation and disappearance of multiple intermetallic phases during accelerated electromigration and takes into account the non-equilibrium formation of vacancies due to electromigration. The work compares nucleation, growth, and coalescence of intermetallic layers during transient liquid phase bonding and virtual joint structure evolution subjected to accelerated electromigration conditions at different temperatures. The changes in the rate of dissolution of Cu from intermetallics and the differences in the evolution of intermetallic layers depending on whether they act as cathodes or anodes are accounted for and are compared favorably with experiments. The model considers non-equilibrium evolution of vacancies that form due to differences in couplings between diffusing atoms and electron flows. This work is significant as the point defect evolution in 3DIC solder joints during electromigration has deep implications to the formation and coalescence of voids that ultimately compromise the structural and functional integrity of the joints.

cond-mat.mtrl-sci↗