arXiv subjects

Diksha Sharma

Publications and source records attributed to Diksha Sharma.

7 recordsLinked to original sources

arXiv2025

Quantum neural networks facilitating quantum state classification

The classification of quantum states into distinct classes poses a significant challenge. In this study, we address this problem using quantum neural networks in combination with a problem-inspired circuit and customised as well as predefined ansätz. To facilitate the resource-efficient quantum state classification, we construct the dataset of quantum states using the proposed problem-inspired circuit. The problem-inspired circuit incorporates two-qubit parameterised unitary gates of varying entangling power, which is further integrated with the ansätz, developing an entire quantum neural network. To demonstrate the capability of the selected ansätz, we visualise the mitigated barren plateaus. The designed quantum neural network demonstrates the efficiency in binary and multi-class classification tasks. This work establishes a foundation for the classification of multi-qubit quantum states and offers the potential for generalisation to multi-qubit pure quantum states.

quant-ph↗

arXiv2024

Harnessing Quantum Support Vector Machines for Cross-Domain Classification of Quantum States

In the present study, we use cross-domain classification using quantum machine learning for quantum advantages to readdress the entanglement versus separability paradigm. The inherent structure of quantum states and its relation to a particular class of quantum states are used to intuitively classify testing states from domains different from training states, called \textit{cross-domain classification}. Using our quantum machine learning algorithm, we demonstrate efficient classifications of two-qubit mixed states into entangled and separable classes. For analyzing the quantumness of correlations, our model adequately classifies Bell diagonal states as zero and non-zero discord states. In addition, we also extend our analysis to evaluate the robustness of our model using random local unitary transformations. Our results demonstrate the potential of the quantum support vector machine for classifying quantum states across the multi-dimensional Hilbert space in comparison to classical support vector machines and neural networks.

quant-ph↗

arXiv2023

Knowledge-based in silico models and dataset for the comparative evaluation of mammography AI for a range of breast characteristics, lesion conspicuities and doses

To generate evidence regarding the safety and efficacy of artificial intelligence (AI) enabled medical devices, AI models need to be evaluated on a diverse population of patient cases, some of which may not be readily available. We propose an evaluation approach for testing medical imaging AI models that relies on in silico imaging pipelines in which stochastic digital models of human anatomy (in object space) with and without pathology are imaged using a digital replica imaging acquisition system to generate realistic synthetic image datasets. Here, we release M-SYNTH, a dataset of cohorts with four breast fibroglandular density distributions imaged at different exposure levels using Monte Carlo x-ray simulations with the publicly available Virtual Imaging Clinical Trial for Regulatory Evaluation (VICTRE) toolkit. We utilize the synthetic dataset to analyze AI model performance and find that model performance decreases with increasing breast density and increases with higher mass density, as expected. As exposure levels decrease, AI model performance drops with the highest performance achieved at exposure levels lower than the nominal recommended dose for the breast type.

eess.IV↗

arXiv2023

Quantum-inspired attribute selection algorithm: A Fidelity-based Quantum Decision Tree

A classical decision tree is completely based on splitting measures, which utilize the occurrence of random events in correspondence to its class labels in order to optimally segregate datasets. However, the splitting measures are based on greedy strategy, which leads to construction of an imbalanced tree and hence decreases the prediction accuracy of the classical decision tree algorithm. An intriguing approach is to utilize the foundational aspects of quantum computing for enhancing decision tree algorithm. Therefore, in this work, we propose to use fidelity as a quantum splitting criterion to construct an efficient and balanced quantum decision tree. For this, we construct a quantum state using the occurrence of random events in a feature and its corresponding class. The quantum state is further utilized to compute fidelity for determining the splitting attribute among all features. Using numerical analysis, our results clearly demonstrate that the proposed algorithm cooperatively ensures the construction of a balanced tree. We further compared the efficiency of our proposed quantum splitting criterion to different classical splitting criteria on balanced and imbalanced datasets. Our simulation results show that the proposed splitting criterion exceeds all classical splitting criteria for all possible evaluation metrics.

quant-ph↗

arXiv2023

The role of entanglement for enhancing the efficiency of quantum kernels towards classification

Quantum kernels are considered as potential resources to illustrate benefits of quantum computing in machine learning. Considering the impact of hyperparameters on the performance of a classical machine learning model, it is imperative to identify promising hyperparameters using quantum kernel methods in order to achieve quantum advantages. In this work, we analyse and classify sentiments of textual data using a new quantum kernel based on linear and full entangled circuits as hyperparameters for controlling the correlation among words. We also find that the use of linear and full entanglement further controls the expressivity of the Quantum Support Vector Machine (QSVM). In addition, we also compare the efficiency of the proposed circuit with other quantum circuits and classical machine learning algorithms. Our results show that the proposed fully entangled circuit outperforms all other fully or linearly entangled circuits in addition to classical algorithms for most of the features. In fact, as the feature increases the efficiency of our proposed fully entangled model also increases significantly.

quant-ph↗

arXiv2022

Impact of Radiation and Slip Conditions on MHD Flow of Nanofluid Past an Exponentially Stretched Surface

The current research establishes magnetohydrodynamics (MHD) boundary layer flow with heat and mass transfer of a nanofluid over an exponentially extending sheet embedded in a porous medium. During this exploration, nanoparticles, single-wall carbon nanotubes (SWCNTs) and multi-wall carbon nanotubes (MWCNTs) are recruited, while lamp fuel oil is being utilised as a base fluid for the diffusion of nano materials. The effects of warm radiation and an inclined magnetic field are included. In addition, rather than no-slip assumptions at the surface, velocity slides as well as thermal upsurge are incorporated in this study. Similarity transformations are implemented to adapt a set of partial differential equations into a system of non-linear ordinary differential equations. The bvp4c solver and Keller-box approach are employed to tackle nonlinear ordinary differential equations numerically. The significance of prominent parameters such as the Darcy Forchheimer model, magnetic field, radiation, suction, velocity slip, and temperature jump is visually probed and addressed in depth. In fact, the evolution of the coefficient of skin friction and percentage of heat shipping (Nusselt number) for both SWCNTs and MWCNTs is presented in tabular form. The temperature goes up as the magnetic parameter rises. Temperature has been seen to be decreased as the thermal slip parameter is improved. The results indicate that SWCNTs yield a higher coefficient of skin friction and speed of heat transformation than MWCNTs.

math.DS↗

arXiv2020

How angry are your customers? Sentiment analysis of support tickets that escalate

Software support ticket escalations can be an extremely costly burden for software organizations all over the world. Consequently, there exists an interest in researching how to better enable support analysts to handle such escalations. In order to do so, we need to develop tools to reliably predict if, and when, a support ticket becomes a candidate for escalation. This paper explores the use of sentiment analysis tools on customer-support analyst conversations to find indicators of when a particular support ticket may be escalated. The results of this research indicate a considerable difference in the sentiment between escalated support tickets and non-escalated support tickets. Thus, this preliminary research provides us with the necessary information to further investigate how we can reliably predict support ticket escalations, and subsequently to provide insight to support analysts to better enable them to handle support tickets that may be escalated.

cs.SE↗