arXiv subjects

Kaustav Chakraborty

Publications and source records attributed to Kaustav Chakraborty.

24 records · Page 2Linked to original sources

arXiv2019

Multiple Style-Transfer in Real-Time

Style transfer aims to combine the content of one image with the artistic style of another. It was discovered that lower levels of convolutional networks captured style information, while higher levels captures content information. The original style transfer formulation used a weighted combination of VGG-16 layer activations to achieve this goal. Later, this was accomplished in real-time using a feed-forward network to learn the optimal combination of style and content features from the respective images. The first aim of our project was to introduce a framework for capturing the style from several images at once. We propose a method that extends the original real-time style transfer formulation by combining the features of several style images. This method successfully captures color information from the separate style images. The other aim of our project was to improve the temporal style continuity from frame to frame. Accordingly, we have experimented with the temporal stability of the output images and discussed the various available techniques that could be employed as alternatives.

cs.CV↗

arXiv2019

Density Power Downweighting and Robust Inference: Some New Strategies

Preserving the robustness of the procedure has, at the present time, become almost a default requirement for statistical data analysis. Since efficiency at the model and robustness under misspecification of the model are often in conflict, it is important to choose such inference procedures which provide the best compromise between these two concepts. Some minimum Bregman divergence estimators and related tests of hypothesis seem to be able to do well in this respect, with the procedures based on the density power divergence providing the existing standard. In this paper we propose a new family of Bregman divergences which is a superfamily encompassing the density power divergence. This paper describes the inference procedures resulting from this new family of divergences, and makes a strong case for the utility of this divergence family in statistical inference.

math.ST↗

arXiv2019

Consequences of $μ$-$τ$ reflection symmetry for $3+1$ neutrino mixing

We investigate the consequences of $μ-τ$ reflection symmetry in presence of a light sterile neutrino for the $3+1$ neutrino mixing scheme. We discuss the implications of total $μ-τ$ reflection symmetry as well partial $μ-τ$ reflection symmetry. For the total $μ-τ$ reflection symmetry we find values of $θ_{23}$ and $δ$ remains confined near $π/4$ and $\pm π/2$ respectively. The current allowed region for $θ_{23}$ and $δ$ in case of inverted hierarchy lies outside the area preferred by the total $μ-τ$ reflection symmetry. However, interesting predictions on the neutrino mixing angles and Dirac CP violating phases are obtained considering partial $μ-τ$ reflection symmetry. We obtain predictive correlations between the neutrino mixing angle $θ_{23}$ and Dirac CP phase $δ$ and study the testability of these correlations at the future long baseline experiment DUNE. We find that while the imposition of $μ-τ$ reflection symmetry in the first column admit both normal and inverted neutrino mass hierarchy, demanding $μ-τ$ reflection symmetry for the second column excludes the inverted hierarchy. Interestingly, the sterile mixing angle $θ_{34}$ gets tightly constrained considering the $μ-τ$ reflection symmetry in the fourth column. We also study consequences of $μ-τ$ reflection symmetry for the Majorana phases and neutrinoless double beta decay.

hep-ph↗

arXiv2019

Enhancing the hierarchy and octant sensitivity of ESS$ν$SB in conjunction with T2K, NO$ν$A and ICAL@INO

The main aim of the ESS$ν$SB proposal is the discovery of the leptonic CP phase $δ_{CP}$ with a high significance ($5σ$ for 50% values of $δ_{CP}$) by utilizing the physics at the second oscillation maxima of the $P_{μe}$ channel. It can achieve $3σ$ sensitivity to hierarchy for all values of $δ_{CP}$. In this work, we concentrate on the hierarchy and octant sensitivity of the ESS$ν$SB experiment. We show that combining the ESS$ν$SB experiment with the atmospheric neutrino data from the proposed India-based Neutrino Observatory(INO) experiment can result in an increased sensitivity to mass hierarchy. In addition, we also combine the results from the ongoing experiments T2K and NO$ν$A assuming their full runtime and present the combined sensitivity of ESS$ν$SB + ICAL@INO + T2K + NO$ν$A. We show that while by itself ESS$ν$SB can have up to $3σ$ hierarchy sensitivity, the combination of all the experiments can give up to $5σ$ sensitivity depending on the true hierarchy-octant combination. The octant sensitivity of ESS$ν$SB is low by itself. However the combined sensitivity of all the above experiments can give up to $3σ$ sensitivity depending on the choice of true hierarchy and octant. We discuss the various degeneracies and the synergies that lead to the enhanced sensitivity when combining different experimental data.

hep-ph↗

arXiv2019

Exploring Partial $μ$-$τ$ Reflection Symmetry at DUNE and Hyper-Kamiokande

We study origin, consequences and testability of a hypothesis of `partial $μ$-$τ$' reflection symmetry. This symmetry predicts $ |U_{μi}|=|U_{τi}|~(i=1,2,3) $ for a single column of the leptonic mixing matrix $U$. Depending on whether this symmetry holds for the first or second column of $U$ different correlations between $θ_{23}$ and $ δ_{CP} $ can be obtained. This symmetry can be obtained using discrete flavour symmetries. In particular, all the subgroups of SU(3) with 3-dimensional irreducible representation which are classified as class C or D can lead to partial $μ$-$τ$ reflection symmetry. We show how the predictions of this symmetry compare with the allowed area in the $\sin^2θ_{23} - δ_{CP}$ plane as obtained from the global analysis of neutrino oscillation data. Furthermore, we study the possibility of testing these symmetries at the proposed DUNE and Hyper-Kamiokande (HK) experiments (T2HK, T2HKK), by incorporating the correlations between $θ_{23}$ and $ δ_{CP}$ predicted by the symmetries. We find that when simulated data of DUNE and HK is fitted with the symmetry predictions, the $θ_{23}-δ_{CP}$ parameter space gets largely restricted near the CP conserving values of $ δ_{CP} $. Finally, we illustrate the capability of these experiments to distinguish between the two cases leading to partial $μ-τ$ symmetry namely $|U_{\mu1}| = |U_{τ1}|$ and $|U_{μ2}| = |U_{τ2}|$.

hep-ph↗

arXiv2017

Spotlighting the sensitivities of T2HK,T2HKK and DUNE

Neutrino oscillation physics has entered the precision era and the potential forthcoming experiments Hyper-Kamiokande and Deep Under-ground Neutrino Experiment (DUNE) are expected to lead this endeavor. In this paper we perform a comprehensive study of the octant, mass hierarchy and CP discovery sensitivities of DUNE, T2HK & T2HKK in their individual capacity and investigate the synergies of the aforementioned experiments with the on going T2K and NO$ν$A experiments. We present a comparative account of the probabilities at the three baselines and explore in detail the physics issues which can cause the discrepancies in the sensitivities among the different experiments. We also find out the optimal exposure required by these experiments for achieving $5σ$ hierarchy and octant sensitivity and to discover CP violation at $3σ$ for 60\% values of $δ_{CP}$. In addition we vary the neutrino-antineutrino runtime ratios for T2HK & T2HKK and check if the sensitivities are affected significantly due to this.

hep-ph↗