Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)

Performance Evaluation of Hybrid Quantum-Classical Algorithms Versus Traditional AI Models

V Divakar Naidu, Dr. A.V.H. Sai Prasad Dr. A.V.H. Sai Prasad, Pentamala Prakash Babu, Mallula Praveen Kumar, Thatiparthy Pavani, Suresh Nalla

Published Oct 8, 2025• 0 citations
Publisher
DOI
10.1109/i-smac65734.2025.11597815

Related papers

Quantum-Classical Performance Benchmark in Nanomaterial Informatics: A Comprehensive Evaluation of Hybrid Quantum Neural Networks versus Deep Learning Architectures

Salah A. M. Elshourbagy

•Jan 1, 2026•0 citations

Hybrid Quantum–Classical Algorithms to Accelerate AI

Bhavya M. Patel

•Jan 1, 2025•0 citations

Benchmarking and Performance Evaluation of Quantum AI Algorithms

Talabattula Divya Sai Sri Akanksha, Chodapuneedi Sowjanya, Manjula Devarakonda Venkata

Quantum AI is a new way to think about the future of Artificial Intelligence through the merging of two technologies: quantum computing and advanced machine learning. Quantum AI has the potential to provide solutions to computationally difficult problems that traditional systems cannot solve. Quantum AI's continued success will rely on advancements in hybrid quantum-classical solutions, as well as the expansion of quantum hardware; hence, research remains necessary to conduct systematic benchmarks as well as performance analysis in order to determine both how well quantum AI algorithms perform practically, and where their limitations lie. In this chapter, detailed examinations of benchmarks that have been developed specifically for quantum AI are presented alongside evaluative metrics across various key areas of performance including algorithmic efficiency, scalability, noise immunity, and resource use for various types of quantum architectures. The chapter integrates dialogue between theory and practice to demonstrate ways in which performance metrics (i.e., quantum circuit depth, fidelity, convergence behaviour and computational complexity) can effectively compare Quantum AI algorithms to classical algorithms. Particular attention is given to hybrid variational algorithms, quantum-enhanced optimization algorithms, and quantum machine learning. Using benchmark datasets, simulation frameworks and new evaluation methods, there are several challenges in terms of reproducibility, hardware variability and algorithm/hardware co-design that emerge through analysis of the above metrics as well. Aside from focusing on a more broad view of how Performance Evaluation relates to developing more effective Quantum AI Solutions. In addition to viewing Benchmarking as a technical exercise solely, this chapter provides a more holistic view in relation to a human perspective of the situation. This paper outlines how performance evaluation practices can be used as a considerable and well-designed means of bridging the gap between the promise of Quantum Technology at an abstract level, and the ability of Quantum Technology to produce tangible outcomes in the real world by adopting a more comprehensive Governance-oriented approach to Benchmarking Quantum AI Systems. The purpose of this chapter is to assist the growing community of researchers, practitioners, and Strategic Technology Professionals by providing them with a clear and structured framework for Evaluation of Quantum AI Algorithms. There is a significant focus of the chapter on establishing important reference points that organizations can use to determine when an intelligent system has successfully achieved a true and validated quantum advantage. The ultimate goal is to assist in guiding the responsible development and implementation of Quantum AI to enable the Potential Benefits of Quantum AI to be realized, in ways that are meaningful and impactful.

Indian Journal of Computer Science•Oct 1, 2026•0 citations

Hybrid Quantum and Classical Machine Learning

Zeph Hardie

Queensland University of Technology eBooks•Jan 1, 2026•0 citations

Quantum AI in Speech Emotion Recognition

Michael Norval, Zenghui Wang

Abstract This study explores Quantum AI’s potential in enhancing Speech Emotion Recognition (SER) systems. Our primary objective is to evaluate the performance of quantum-inspired algorithms compared to classical machine-learning approaches in accurately identifying and classifying emotions from speech signals. We hypothesise that quantum computing techniques can improve the efficiency and accuracy of emotion recognition, particularly in handling complex, highdimensional acoustic data. We developed a hybrid quantum-classical model that combines quantum-inspired neural networks with traditional feature extraction techniques. Our approach utilises quantum circuits for data encoding and quantum variational algorithms for classification. We implemented quantum versions of support vector machines (QSVM) and quantum approximate optimisation algorithms (QAOA) for emotion classification tasks. These quantum methods were compared against a hybrid Convolutional Neural Network and Long Short Term (LSTM). The hybrid network is called a Convolutional Long Short Term network(CLSTM). The study employed a custom dataset for speech emotion recognition. We prepossessed the audio data to extract relevant acoustic features, including mel-frequency cepstral coefficients (MFCCs), pitch, and energy. In contrast, our implemented quantum-inspired model for this corpus showed lower performance, with the highest training accuracy of 30%, struggling with most emotions but performing best with ’Trust’. These varied results across different implementations suggest that while Quantum AI techniques show promise in advancing speech emotion recognition, their effectiveness may depend on the specific dataset, language, and model architecture used. Finally, the speed of a simulated quantum network is shown to outperform current methods in terms of matrix multiplication by leaps and bounds.

Entropy•Nov 26, 2025•5 citations

An Intelligent Hybrid Cryptographic Security Framework: Leveraging Artificial Intelligence to Defend Critical Systems Against Next-Generation Quantum Computing Attacks

Gulshan Verma

Quantum computing advances rapidly and brings significant risks to traditional methods of encrypting data. As a result, new and innovative methods must be designed to provide advanced security techniques that will defend against both classical and quantum-based attacks. This work explores an AI-assisted hybrid cryptographic architecture (HCA) that combines traditional encryption, post-quantum methods of data encryption, and artificial intelligence to create a more secure system than offered by any of the individual methods alone. The HCA uses Algorithm 1, which applies an adaptive key management and threat detection protocol utilizing classical and lattice-based cryptographic techniques to create a unified hybrid key that provides dual layers of security. The HCA uses AI-assisted anomaly detection to monitor system functions, identify threats, and adjust security models accordingly. The HCA employs efficient key creation, encrypting, and decrypting processes with lower computational requirements and optimized resource usage. The performance of the HCA is evaluated against several performance metrics consisting of security level, computational complexity, total memory utilized, energy consumed, scalability, accuracy of detection, and false positive rate. The performance metrics demonstrate that the HCA significantly outperforms existing methods by achieving higher levels of security, shorter amounts of time to process information, and greater quality of detection while generating fewer false positives. The integration of AI into the HCA allows for greater adaptation to new and evolving cyber threats to respond in real time.As confirmed by both tables and graphical diagrams, the framework's proposed solution successfully balances security, efficiency, and scalability. This places it in a position to be used in today's climate of modern technologies (cloud computing, IoT networks, and secure communications). Overall, this study offers an effective and future-oriented way to deal with the obstacles of cyberspace security presented by quantum computing.

INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT•Apr 1, 2026•0 citations
Quantum AI is a new way to think about the future of Artificial Intelligence through the merging of two technologies: quantum computing and advanced machine learning. Quantum AI has the potential to provide solutions to computationally difficult problems that traditional systems cannot solve. Quantum AI's continued success will rely on advancements in hybrid quantum-classical solutions, as well as the expansion of quantum hardware; hence, research remains necessary to conduct systematic benchmarks as well as performance analysis in order to determine both how well quantum AI algorithms perform practically, and where their limitations lie. In this chapter, detailed examinations of benchmarks that have been developed specifically for quantum AI are presented alongside evaluative metrics across various key areas of performance including algorithmic efficiency, scalability, noise immunity, and resource use for various types of quantum architectures. The chapter integrates dialogue between theory and practice to demonstrate ways in which performance metrics (i.e., quantum circuit depth, fidelity, convergence behaviour and computational complexity) can effectively compare Quantum AI algorithms to classical algorithms. Particular attention is given to hybrid variational algorithms, quantum-enhanced optimization algorithms, and quantum machine learning. Using benchmark datasets, simulation frameworks and new evaluation methods, there are several challenges in terms of reproducibility, hardware variability and algorithm/hardware co-design that emerge through analysis of the above metrics as well. Aside from focusing on a more broad view of how Performance Evaluation relates to developing more effective Quantum AI Solutions. In addition to viewing Benchmarking as a technical exercise solely, this chapter provides a more holistic view in relation to a human perspective of the situation. This paper outlines how performance evaluation practices can be used as a considerable and well-designed means of bridging the gap between the promise of Quantum Technology at an abstract level, and the ability of Quantum Technology to produce tangible outcomes in the real world by adopting a more comprehensive Governance-oriented approach to Benchmarking Quantum AI Systems. The purpose of this chapter is to assist the growing community of researchers, practitioners, and Strategic Technology Professionals by providing them with a clear and structured framework for Evaluation of Quantum AI Algorithms. There is a significant focus of the chapter on establishing important reference points that organizations can use to determine when an intelligent system has successfully achieved a true and validated quantum advantage. The ultimate goal is to assist in guiding the responsible development and implementation of Quantum AI to enable the Potential Benefits of Quantum AI to be realized, in ways that are meaningful and impactful.
Abstract This study explores Quantum AI’s potential in enhancing Speech Emotion Recognition (SER) systems. Our primary objective is to evaluate the performance of quantum-inspired algorithms compared to classical machine-learning approaches in accurately identifying and classifying emotions from speech signals. We hypothesise that quantum computing techniques can improve the efficiency and accuracy of emotion recognition, particularly in handling complex, highdimensional acoustic data. We developed a hybrid quantum-classical model that combines quantum-inspired neural networks with traditional feature extraction techniques. Our approach utilises quantum circuits for data encoding and quantum variational algorithms for classification. We implemented quantum versions of support vector machines (QSVM) and quantum approximate optimisation algorithms (QAOA) for emotion classification tasks. These quantum methods were compared against a hybrid Convolutional Neural Network and Long Short Term (LSTM). The hybrid network is called a Convolutional Long Short Term network(CLSTM). The study employed a custom dataset for speech emotion recognition. We prepossessed the audio data to extract relevant acoustic features, including mel-frequency cepstral coefficients (MFCCs), pitch, and energy. In contrast, our implemented quantum-inspired model for this corpus showed lower performance, with the highest training accuracy of 30%, struggling with most emotions but performing best with ’Trust’. These varied results across different implementations suggest that while Quantum AI techniques show promise in advancing speech emotion recognition, their effectiveness may depend on the specific dataset, language, and model architecture used. Finally, the speed of a simulated quantum network is shown to outperform current methods in terms of matrix multiplication by leaps and bounds.
Quantum computing advances rapidly and brings significant risks to traditional methods of encrypting data. As a result, new and innovative methods must be designed to provide advanced security techniques that will defend against both classical and quantum-based attacks. This work explores an AI-assisted hybrid cryptographic architecture (HCA) that combines traditional encryption, post-quantum methods of data encryption, and artificial intelligence to create a more secure system than offered by any of the individual methods alone. The HCA uses Algorithm 1, which applies an adaptive key management and threat detection protocol utilizing classical and lattice-based cryptographic techniques to create a unified hybrid key that provides dual layers of security. The HCA uses AI-assisted anomaly detection to monitor system functions, identify threats, and adjust security models accordingly. The HCA employs efficient key creation, encrypting, and decrypting processes with lower computational requirements and optimized resource usage. The performance of the HCA is evaluated against several performance metrics consisting of security level, computational complexity, total memory utilized, energy consumed, scalability, accuracy of detection, and false positive rate. The performance metrics demonstrate that the HCA significantly outperforms existing methods by achieving higher levels of security, shorter amounts of time to process information, and greater quality of detection while generating fewer false positives. The integration of AI into the HCA allows for greater adaptation to new and evolving cyber threats to respond in real time.As confirmed by both tables and graphical diagrams, the framework's proposed solution successfully balances security, efficiency, and scalability. This places it in a position to be used in today's climate of modern technologies (cloud computing, IoT networks, and secure communications). Overall, this study offers an effective and future-oriented way to deal with the obstacles of cyberspace security presented by quantum computing.