Advances in Nonlinear Dynamics, Volume I (ICNDA 2023)
This paper presents an active learning framework for probabilistic machine learning models of dynamical systems, enhancing sample efficiency and model accuracy while providing uncertainty quantification.
Chaos, Solitons & Fractals, 2023
This study employs Physics-Informed Neural Networks to discover multi-valley dark soliton solutions in the multi-component Manakov model, demonstrating the effectiveness of data-driven approaches for complex nonlinear systems.
Romanian Reports in Physics
A comprehensive introduction to solving differential equations using various machine learning approaches, including neural networks, Gaussian processes, and physics-informed methods.
Scientific Reports, 2021
This work demonstrates the application of Bayesian optimization techniques to efficiently tune experimental parameters for Bose-Einstein condensate production, significantly reducing experimental overhead.
INFITT
This paper presents novel spell-checking approaches for Tamil language using SymSpell algorithm and LSTM neural networks, addressing the challenges of morphologically rich languages.
Nonlinear Dynamics, 2019
An investigation of soliton dynamics in the fourth-order nonlocal nonlinear Schrödinger equation, revealing novel behaviors and stability properties of these nonlinear wave structures.
Preprint, 2018
A theoretical study of analytical signatures in ultracold atomic systems, providing new insights into quantum many-body physics and potential experimental observables.
Optical Materials, 2020
A comprehensive characterization of a novel semi-organic crystal, Brucinium Di-Hydrogen Borate Hydrate, revealing its potential for microelectronics and high-power laser applications.
Journal of Molecular Structure, 2019
Investigation of Brucinium Bromide Hydrate crystal properties for applications in optical parametric oscillators, high-power lasers, piezo-sensors and transducers.