arXiv subjects

Priyam Das

Publications and source records attributed to Priyam Das.

At least 37 records · Page 2Linked to original sources

arXiv2020

On Second order correctness of Bootstrap in Logistic Regression

In the fields of clinical trials, biomedical surveys, marketing, banking, with dichotomous response variable, the logistic regression is considered as an alternative convenient approach to linear regression. In this paper, we develop a novel bootstrap technique based on perturbation resampling method for approximating the distribution of the maximum likelihood estimator (MLE) of the regression parameter vector. We establish second order correctness of the proposed bootstrap method after proper studentization and smoothing. It is shown that inferences drawn based on the proposed bootstrap method are more accurate compared to that based on asymptotic normality. The main challenge in establishing second order correctness remains in the fact that the response variable being binary, the resulting MLE has a lattice structure. We show the direct bootstrapping approach fails even after studentization. We adopt smoothing technique developed in Lahiri (1993) to ensure that the smoothed studentized version of the MLE has a density. Similar smoothing strategy is employed to the bootstrap version also to achieve second order correct approximation.

math.ST↗

arXiv2020

Formation of Solitonic Bound State via Light-Matter Interaction

Exchange of energy by means of light-matter interaction provides a new dimension to various nonlinear dynamical systems. Here, the effects of light-matter interaction are investigated for a situation, where two counter-propagating, orthogonally polarized laser pulses are incident on the atomic condensate. It's observed that a localized laser pulse profile can induce localized modes in Bose-Einstein condensate. A stability analysis performed using Vakhitov-Kolokolov-like criterion has established that these localized modes are stable, when the atom-atom interaction is repulsive. The cooperative effects of light-matter interactions and atom-atom interactions on the Lieb-mode have been studied in the stable region through atomic dispersion, revealing the signature of bound state formation when the optical potential is Pöschl-Teller type. The energy diagram also indicates a continuous transfer of energy from the laser pulses to the atoms as the light-matter interaction changes its sign.

cond-mat.quant-gas↗

arXiv2019

A distribution-free smoothed combination method of biomarkers to improve diagnostic accuracy in multi-category classification

Results from multiple diagnostic tests are usually combined to improve the overall diagnostic accuracy. For binary classification, maximization of the empirical estimate of the area under the receiver operating characteristic (ROC) curve is widely adopted to produce the optimal linear combination of multiple biomarkers. In the presence of large number of biomarkers, this method proves to be computationally expensive and difficult to implement since it involves maximization of a discontinuous, non-smooth function for which gradient-based methods cannot be used directly. Complexity of this problem increases when the classification problem becomes multi-category. In this article, we develop a linear combination method that maximizes a smooth approximation of the empirical Hypervolume Under Manifolds (HUM) for multi-category outcome. We approximate HUM by replacing the indicator function with the sigmoid function or normal cumulative distribution function (CDF). With the above smooth approximations, efficient gradient-based algorithms can be employed to obtain better solution with less computing time. We show that under some regularity conditions, the proposed method yields consistent estimates of the coefficient parameters. We also derive the asymptotic normality of the coefficient estimates. We conduct extensive simulations to examine our methods. Under different simulation scenarios, the proposed methods are compared with other existing methods and are shown to outperform them in terms of diagnostic accuracy. The proposed method is illustrated using two real medical data sets.

stat.ME↗

arXiv2019

Recursive Modified Pattern Search on High-dimensional Simplex : A Blackbox Optimization Technique

In this paper, a novel derivative-free pattern search based algorithm for Black-box optimization is proposed over a simplex constrained parameter space. At each iteration, starting from the current solution, new possible set of solutions are found by adding a set of derived step-size vectors to the initial starting point. While deriving these step-size vectors, precautions and adjustments are considered so that the set of new possible solution points still remain within the simplex constrained space. Thus, no extra time is spent in evaluating the (possibly expensive) objective function at infeasible points (points outside the unit-simplex space). While minimizing any objective function of m parameters, within each iteration, the objective function is evaluated at 2m new possible solution points. So, upto 2m parallel threads can be incorporated which makes the computation even faster while optimizing expensive objective functions over high-dimensional parameter space. Once a local minimum is discovered, in order to find a better solution, a novel `re-start' strategy is considered to increase the likelihood of finding a better solution. Unlike existing pattern search based methods, a sparsity control parameter is introduced which can be used to induce sparsity in the solution in case the solution is expected to be sparse in prior. A comparative study of the performances of the proposed algorithm and other existing algorithms are shown for a few low, moderate and high-dimensional optimization problems. Upto 338 folds improvement in computation time is achieved using the proposed algorithm over Genetic algorithm along with better solution. The proposed algorithm is used to estimate the simultaneous quantiles of North Atlantic Hurricane velocities during 1981-2006 by maximizing a non-closed form likelihood function with (possibly) multiple maximums.

math.OC↗

arXiv2018

NExUS: Bayesian simultaneous network estimation across unequal sample sizes

Network-based analyses of high-throughput genomics data provide a holistic, systems-level understanding of various biological mechanisms for a common population. However, when estimating multiple networks across heterogeneous sub-populations, varying sample sizes pose a challenge in the estimation and inference, as network differences may be driven by differences in power. We are particularly interested in addressing this challenge in the context of proteomic networks for related cancers, as the number of subjects available for rare cancer (sub-)types is often limited. We develop NExUS (Network Estimation across Unequal Sample sizes), a Bayesian method that enables joint learning of multiple networks while avoiding artefactual relationship between sample size and network sparsity. We demonstrate through simulations that NExUS outperforms existing network estimation methods in this context, and apply it to learn network similarity and shared pathway activity for groups of cancers with related origins represented in The Cancer Genome Atlas (TCGA) proteomic data.

stat.AP↗

arXiv2018

Chemical Oscillation in Ultracold Chemistry

We demonstrate the occurrence of oscillatory reactions in the ultra-cold chemistry of atom-molecular Bose-Einstein condensate. Nonlinear oscillations in the mean-field dynamics occur for a specific range of elliptic modulus, giving rise to both in- and out-phase modulations in the atom-molecule population density. The reaction front velocity is found to be controlled by photoassociation, which also regulates the condensate density. Two distinct pair of in-phase bright localized gap solitons are found as exact solutions, existence of one of which necessarily requires a background. Cnoidal atomic density-waves in a plane wave molecular background are observed in both attractive and repulsive domains. Role of intra- and inter-species interactions on both existence and stability is explicated in the presence of photoassociation.

nlin.PS↗

arXiv2017

Quantum State Transfer through Coherent Atom-Molecule Conversion in Bose-Einstein Condensate

We demonstrate complete quantum state transfer of an atomic Bose-Einstein condensate to molecular condensate, mediated by solitonic excitations in a cigar shaped mean-field geometry. Starting with a localized solitonic atomic condensate, we show compatible gray solitonic configuration in the molecular condensate, which results in complete atom-molecule conversion. The effect of inter and intra-species interactions on the formation of molecular condensate is explicated in the presence of Raman Photoassociation. It is found that photoassociation plays a crucial role in the coherent atom-molecule conversion as well as in the soliton dynamics. The gray soliton dispersion reveals bistable behaviour, showing a re-entrant phase in a physically accessible parametric domain.

cond-mat.quant-gas↗

arXiv2017

Lattice induced stripe phase in Bose-Einstein condensate under non-inertial and inertial motion

We consider a parametrically forced Bose-Einstein condensate in the combined presence of an optical lattice and harmonic oscillator potential in the mean field approach. A spatial symmetry broken Bose-condensed phase in non-inertial and inertial frame yields a stripe phase in the presence of both cubic and quintic nonlinearities. We show that existence of such stripe phase solely depends on the interplay between the quintic nonlinearity and lattice potential. Furthermore, we observe that a time-dependent harmonic oscillator frequency destroys such stripe ordering. A linear stability analysis of the obtained solution is performed and we found that the solution is stable. In order to gain a better understanding of the underlying physics, we compute the energy, showing nonlinear compression of the condensate in some parameter domain.

cond-mat.quant-gas↗

arXiv2016

Analyzing Ozone Concentration by Bayesian Spatio-temporal Quantile Regression

Ground level Ozone is one of the six common air-pollutants on which the EPA has set national air quality standards. In order to capture the spatio-temporal trend of 1-hour and 8-hour average ozone concentration in the US, we develop a method for spatio-temporal simultaneous quantile regression. Unlike existing procedures, in the proposed method, smoothing across the sites is incorporated within modeling assumptions thus allowing borrowing of information across locations, an essential step when the number of samples in each location is low. The quantile function has been assumed to be linear in time and smooth over space and at any given site is given by a convex combination of two monotone increasing functions $ξ_1$ and $ξ_2$ not depending on time. A B-spline basis expansion with increasing coefficients varying smoothly over the space is used to put a prior and a Bayesian analysis is performed. We analyze the average daily 1-hour maximum and 8-hour maximum ozone concentration level data of US and California during 2006-2015 using the proposed method. It is noted that in the last ten years, there is an overall decreasing trend in both 1-hour maximum and 8-hour maximum ozone concentration level over the most parts of the US. In California, an overall a decreasing trend of 1-hour maximum ozone level is observed while no particular overall trend has been observed in the case of 8-hour maximum ozone level.

stat.ME↗

arXiv2016

Bayesian Non-parametric Simultaneous Quantile Regression for Complete and Grid Data

In this paper, we consider Bayesian methods for non-parametric quantile regressions with multiple continuous predictors ranging values in the unit interval. In the first method, the quantile function is assumed to be smooth over the explanatory variable and is expanded in tensor product of B-spline basis functions. While in the second method, the distribution function is assumed to be smooth over the explanatory variable and is expanded in tensor product of B-spline basis functions. Unlike other existing methods of non-parametric quantile regressions, the proposed methods estimate the whole quantile function instead of estimating on a grid of quantiles. Priors on the B-spline coefficients are put in such a way that the monotonicity of the estimated quantile levels are maintained unlike local polynomial quantile regression methods. The proposed methods have also been modified for quantile grid data where only the percentile range of each response observations are known. Simulations studies have been provided for both complete and quantile grid data. The proposed method has been used to estimate the quantiles of US household income data and North Atlantic hurricane intensity data.

stat.ME↗

arXiv2016

Bayesian Quantile Regression Using Random B-spline Series Prior

We consider a Bayesian method for simultaneous quantile regression on a real variable. By monotone transformation, we can make both the response variable and the predictor variable take values in the unit interval. A representation of quantile function is given by a convex combination of two monotone increasing functions $ξ_1$ and $ξ_2$ not depending on the prediction variables. In a Bayesian approach, a prior is put on quantile functions by putting prior distributions on $ξ_1$ and $ξ_2$. The monotonicity constraint on the curves $ξ_1$ and $ξ_2$ are obtained through a spline basis expansion with coefficients increasing and lying in the unit interval. We put a Dirichlet prior distribution on the spacings of the coefficient vector. A finite random series based on splines obeys the shape restrictions. We compare our approach with a Bayesian method using Gaussian process prior through an extensive simulation study and some other Bayesian approaches proposed in the literature. An application to a data on hurricane activities in the Atlantic region is given. We also apply our method on region-wise population data of USA for the period 1985--2010.

stat.ME↗

arXiv2016

Collectively induced many-vortices topology via rotatory Dicke quantum phase transition

We examine the superradiance of a Bose-Einstein condensate pumped with a Laguerre-Gaussian laser of high winding number, e.g., $\ell = 7$. The laser beam transfers its orbital angular momentum (OAM) to the condensate at once due to the collectivity of the superradiance. An $\ell$-fold rotational symmetric structure emerges with the take place of rotatory superradiance. $\ell$ number of single-charge vortices appear at the arms of this structure. Even though the pump and the condensate profiles initially have cylindrical symmetry, we observe that it is broken to $\ell$-fold rotational symmetry during the superradiance. Breaking of the cylindrical symmetry into the $\ell$-fold symmetry and OAM transfer to the BEC become possible after the same critical pump strength. Reorganization of the condensate resembles the ordering in the experiment by Esslinger and colleagues [Nature, {\bf 264}, 1301 (2010)]. We show that the critical point for the onset of the reorganization follows the form of the Dicke quantum phase transition.

cond-mat.quant-gas↗

arXiv2015

Controlled Generation of Nonlinear Resonances through Sinusoidal Lattice Modes in Bose-Einstein Condensate

We study Bose-Einstein condensate in the combined presence of time modulated optical lattice and harmonic trap in the mean-field approach. Through the self-similar method, we show the existence of sinusoidal lattice modes in this inhomogeneous system, commensurate with the lattice potential. A significant advantage of this system is wide tunability of the parameters through chirp management. The combined effect of the interaction, harmonic trap and lattice potential leads to the generation of nonlinear resonances, exactly where the matter wave changes its direction. When the harmonic trap is switched off, the BEC undergoes a nonlinear compression for the static optical lattice potential. For better understanding of chirp management and the nature of the sinusoidal excitation, we investigate the energy spectrum of the condensate, which clearly reveals the generation of nonlinear resonances in the appropriate regime. We have also identified a classical dynamical phase transition occurring in the system, where loss of superfluidity takes the superfluid phase to an insulating state.

cond-mat.quant-gas↗

arXiv2015

Topology of Quantum Grey Soliton in Multi-Component Inhomogeneous Bose-Einstein Condensates

We study the dispersion mechanism of Lieb mode excitations of both single and multi component ultra-cold atomic Bose gas, subject to a harmonic confinement through chirp management. It is shown that in some parameter domain, the hole-like excitations lead to the soliton's negative mass regime, arising due to the coupling between chirp momentum and Kohn mode. In low momenta region the trap considerably affects the dispersion of the grey soliton, which opens a new window to observe Lieb-mode excitations. Further, we extend our analysis to binary condensate, which yields usual shape compatible grey-bright soliton pairs. The inter-species interaction induces a shift in the Lieb-mode excitations, where the pair can form a bound state. We emphasize that the present model provides an opportunity to study such excitations in the low momenta regime, as well as the formation of bound state in binary condensate.

cond-mat.quant-gas↗

arXiv2015

Chirp Control of Sinusoidal Lattice Modes in Bose-Einstein Condensate

A parametrically forced Bose-Einstein condensate (BEC) is studied in the mean field approach for the case of a general moving optical lattice. The interaction between the atoms in the condensate and the time dependent lattice potential leads to a novel propagating superfluid matter wave, which can be controlled through chirp management. This system, when placed in a trap, accelerates and undergoes rapid nonlinear compression, controlled by the chirp. The density achieves its maximum, precisely when the matter wave changes direction. A dynamical phase transition is identified, which takes the superfluid phase to an insulating state. The exact expression for energy is obtained and analyzed in detail to gain physical understanding of the chirp management of the sinusoidal excitations and also the dynamical phase transition.

cond-mat.quant-gas↗

arXiv2014

Realization of the driven nonlinear Schrödinger equation with stationary light

We introduce a versatile platform for studying nonlinear out-of-equilibrium physics. The platform is based on a slow light setup where an optical waveguide is interfaced with cold atoms to realize the driven nonlinear Schrödinger equation with a potential. We compare the proposed setup with similar setups using Bose-Einstein condensates and investigate the system's response under coherent driving for a lattice potential. The slow light setup provides novel angles in the study of nonlinear dynamics due to its advantages in introducing and modulating the driving, the extra tunability over the sign and strength of the available nonlinearities, and the possibility to electromagnetically carve out the underlying potential on demand.

quant-ph↗

arXiv2014

Probing the topological properties of the Jackiw-Rebbi model with light

The Jackiw-Rebbi model describes a one-dimensional Dirac particle coupled to a soliton field and can be equivalently thought of as the model describing a Dirac particle under a Lorentz scalar potential. Neglecting the dynamics of the soliton field, a kink in the background soliton profile yields a topologically protected zero-energy mode for the particle, which in turn leads to charge fractionalization. We show here that the model can be realized in a driven slow-light setup, where photons mimic the Dirac particles and the soliton field can be implemented-and tuned-by adjusting optical parameters such as the atom-photon detuning. Furthermore, we discuss how the existence of the zero-mode, and its topological stability, can be probed naturally by analyzing the transmission spectrum. We conclude by doing an analysis of the robustness of our approach against possible experimental errors in engineering the Jackiw-Rebbi Hamiltonian in this optical set up.

quant-ph↗

arXiv2011

New cross-phase modulated localized solitons in coupled atomic-molecular BEC

The interacting atom-molecule BEC (AMBEC) dynamics is investigated in the mean field ap- proach. The presence of atom-atom, atom-molecule and molecule-molecule interactions, coupled with a characteristically different interaction representing atom-molecule interconversion, endows this system with nonlinearities, which differ significantly from the standard Gross-Pitaevskii (GP) equation. Exact localized solutions are found to belong to two distinct classes. The first ones are analogous to the soliton solutions of the weakly coupled GP equation, whereas the second non- equivalent class is related to the solitons of the strongly coupled BEC. Distinct parameter domains characterize these solitons, some of which are analogous to the complex profile Bloch solitons in magnetic systems. These localized solutions are found to represent a variety of phenomena, which include co-existence of both atom-molecule complex and miscible-immiscible phases. Numerical sta- bility is explicitly checked, as also the stability analysis based on the study of quantum uctuations around our solutions. We also find out the domain of modulation instability in this system.

cond-mat.quant-gas↗