arXiv subjects

Kisor Kumar Sahu

Publications and source records attributed to Kisor Kumar Sahu.

3 recordsLinked to original sources

arXiv2026

Numerically optimized FROG results for the study of red-shifted spectra in multi-frequency Raman generation

When multifrequency Raman scattering is driven in the transient regime by two chirped pump pulses, the resulting anti-Stokes orders exhibit asymmetric spectral broadening toward lower frequencies, leading to a characteristic double-peaked structure in each order. In this Letter, frequency-resolved optical gating (FROG) is used to investigate the spectral evolution of the first anti-Stokes Raman component. To interpret the observed features, we introduce a double-pulse interference model and employ an adaptive learning-based reconstruction algorithm using the Adam optimizer to retrieve the temporal field evolution. The simulation results show good agreement with the experimental measurements. Our analysis indicates that the observed red-shifted spectral component originates from linear Raman processes within the two-photon dressed-state framework.

physics.optics↗

arXiv2020

Analysis of COVID19 Outbreak in India using SEIR model

The prediction of spread patterns of COVID19 virus in India is very difficult due to its versatile demographic as well as meteorological data distribution. Various researchers across the globe have attempted to correlate the interdependency of these data with the spread pattern of COVID19 cases in India. But it is hard to predict the exact pattern, especially the peak in the number of active cases. In the present article we have tried to predict the number of active, recovered, death and total cases of COVID19 in India using generalized SEIR model. In our prediction, the occurrence of peak in the active cases curve has a very close match with the peak in the real data (difference of only one week). Although the number of predicted cases differs with the real number of cases (due to unlocking the movement restrictions gradually from June 2020 onwards), the close resemblance in the actual and predicted time (in the peak of active cases curve) makes this model relatively suitable for analysis of COVID19 outbreak in India.

physics.soc-ph↗

arXiv2020

A new nature inspired modularity function adapted for unsupervised learning involving spatially embedded networks: A comparative analysis

Unsupervised machine learning methods can be of great help in many traditional engineering disciplines, where huge amount of labeled data is not readily available or is extremely difficult or costly to generate. Two specific examples include the structure of granular materials and atomic structure of metallic glasses. While the former is critically important for several hundreds of billion dollars global industries, the latter is still a big puzzle in fundamental science. One thing is common in both the examples is that the particles are the elements of the ensembles that are embedded in Euclidean space and one can create a spatially embedded network to represent their key features. Some recent studies show that clustering, which generically refers to unsupervised learning, holds great promise in partitioning these networks. In many complex networks, the spatial information of nodes play very important role in determining the network properties. So understanding the structure of such networks is very crucial. We have compared the performance of our newly developed modularity function with some of the well-known modularity functions. We performed this comparison by finding the best partition in 2D and 3D granular assemblies. We show that for the class of networks considered in this article, our method produce much better results than the competing methods.

cs.LG↗