The integration of metaheuristics with machine learning methodologies presents significant advantages, particularly in optimization and computational intelligence. This amalgamation leverages the global search capabilities of metaheuristics alongside the pattern recognition and predictive prowess of machine learning, facilitating enhanced convergence rates and solution quality in complex problem spaces. The Quantum Long Short-Term Memory (QLSTM) emerges as a highly efficient deep learning model tailored to tackle such intricate engineering problems. The QLSTM's architecture, comprising data encoding, variational, and quantum measurement layers, facilitates the effective encoding and processing of civil engineering data, leading to heightened prediction accuracy. However, the task of determining optimal values for QLSTM parameters presents challenges due to its NP-problem nature and time-consuming characteristics. To address this, we propose an alternative technique to optimize the QLSTM based on a modified Electric Eel Foraging Optimization (MEEFO). The MEEFO is a modified version of the original EEFO that applies triangular mutation operators to boost the search capability of the traditional EEFO. Thus, the MEEFO optimizes the QLSTM and boosts its prediction performance. To validate the efficacy of our proposed method, we conduct comprehensive experiments utilizing five real-world engineering datasets related to construction and structure engineering. The evaluation outcomes unequivocally demonstrate that the MMEFO significantly enhances the performance of the QLSTM.
The integration of metaheuristics with machine learning methodologies presents significant advantages, particularly in optimization and computational intelligence. This amalgamation leverages the global search capabilities of metaheuristics alongside the pattern recognition and predictive prowess of machine learning, facilitating enhanced convergence rates and solution quality in complex problem spaces. The Quantum Long Short-Term Memory (QLSTM) emerges as a highly efficient deep learning model tailored to tackle such intricate engineering problems. The QLSTM's architecture, comprising data encoding, variational, and quantum measurement layers, facilitates the effective encoding and processing of civil engineering data, leading to heightened prediction accuracy. However, the task of determining optimal values for QLSTM parameters presents challenges due to its NP-problem nature and time-consuming characteristics. To address this, we propose an alternative technique to optimize the QLSTM based on a modified Electric Eel Foraging Optimization (MEEFO). The MEEFO is a modified version of the original EEFO that applies triangular mutation operators to boost the search capability of the traditional EEFO. Thus, the MEEFO optimizes the QLSTM and boosts its prediction performance. To validate the efficacy of our proposed method, we conduct comprehensive experiments utilizing five real-world engineering datasets related to construction and structure engineering. The evaluation outcomes unequivocally demonstrate that the MMEFO significantly enhances the performance of the QLSTM.
In recent years, the electricity consumption in residential sector has witnessed a significant increase due to the population growth on one hand and the widespread adoption of electrical appliances on the other hand.Hence, finding solutions to decrease electricity consumption has become a matter of great interest for researchers.To address this challenge, we propose a Quantum Cognitive IoT (QCIoT) Framework that integrates quantum deep learning with edge computing to optimize energy use in smart homes.Our key innovation is a hybrid Quantum Long Short-Term Memory (QLSTM) model, which enhances traditional LSTM networks by leveraging quantum circuits for improved timeseries forecasting.Specifically, QLSTM employs parameterized quantum gates to process temporal dependencies more efficiently, enabling higher accuracy than classical approaches.We evaluate quantum-enhanced LSTMs (QLSTMs) against classical LSTM baselines on multivariate time-series forecasting.Experimental results demonstrate that QLSTMs significantly outperform classical counterparts, with the multivariate QLSTM (MQLSTM) achieving a 25.8% reduction in RMSE and improving explanatory power by 78.3%.While QLSTMs exhibit slightly slower convergence, they deliver superior generalization, evidenced by lower test loss and stable training dynamics.These advantages stem from quantum parallelism, entanglement and optimized state representation, which enable superior handling of noisy, high-dimensional smart home data.By integrating quantum-enhanced forecasting with edge-based IoT systems, our framework offers a scalable solution for real-time energy management in smart homes.This work bridges quantum computing and smart infrastructure, demonstrating practical benefits for sustainability and energy savings.
Quantum computing combined with machine learning (ML) is a highly promising research area, with numerous studies demonstrating that quantum machine learning (QML) is expected to solve scientific problems more effectively than classical ML. In this work, we present Quantum Long Short-Term Memory (QLSTM), a QML architecture, and demonstrate its effectiveness in drug discovery. We evaluate QLSTM on five benchmark datasets (BBBP, BACE, SIDER, BCAP37, T-47D), and observe consistent performance gains over classical LSTM, with ROC-AUC improvements ranging from 3% to over 6%. Furthermore, QLSTM exhibits improved predictive accuracy as the number of qubits increases, and faster convergence than classical LSTM under the same training conditions. Notably, QLSTM maintains strong robustness against quantum computer noise, outperforming noise-free classical LSTM in certain settings. These findings highlight the potential of QLSTM as a scalable and noise-resilient model for scientific applications, particularly as quantum hardware continues to advance in qubit capacity and fidelity.
Abstract Quantum computing combined with machine learning (ML) is a highly promising research area, with numerous studies demonstrating that quantum machine learning (QML) is expected to solve scientific problems more effectively than classical ML. In this work, we present Quantum Long Short-Term Memory (QLSTM), a QML architecture, and demonstrate its effectiveness in drug discovery. We evaluate QLSTM on five benchmark datasets (BBBP, BACE, SIDER, BCAP37, T-47D), and observe consistent performance gains over classical LSTM, with ROC-AUC improvements ranging from 3% to over 6%. Furthermore, QLSTM exhibits improved predictive accuracy as the number of qubits increases, and faster convergence than classical LSTM under the same training conditions. Notably, QLSTM maintains strong robustness against quantum computer noise, outperforming noise-free classical LSTM in certain settings. These findings highlight the potential of QLSTM as a scalable and noise-resilient model for scientific applications, particularly as quantum hardware continues to advance in qubit capacity and fidelity.
Quantum machine learning (QML) offers new perspectives for modeling complex sequential data by leveraging the computational advantages of quantum computing. In this work, we explore the application of Quantum Long Short-Term Memory (QLSTM) networks for sequence prediction tasks. By integrating parameterized quantum circuits within the classical LSTM architecture, QLSTM models exploit quantum parallelism and entanglement to learn temporal dependencies in data more efficiently. We evaluate the performance of QLSTM on synthetic and real-world sequence prediction tasks and compare it with its classical counterpart. The results demonstrate that QLSTM models achieve competitive performance with fewer trainable parameters, indicating their potential as lightweight and efficient models for sequence learning and prediction.
The growing complexity of urban transportation networks demands intelligent, data-driven systems capable of real-time perception, optimization, and prediction. Within the Internet of intelligent Vehicles (IoIV), optimal sensor placement is essential for enhancing distributed edge intelligence, as it significantly improves traffic observability, ensures accurate data collection, and enables dynamic decision-making. In this context, we propose a dual approach for hybrid quantum-classical (HQC) optimization—classical for quantum and quantum for classical. First, we formulate the optimal sensor placement problem within the quantum approximate optimization algorithm (QAOA) framework using a minimum vertex cover approach to address its NP-hard nature. Moreover, we enhance the QAOA by integrating long short-term memory (LSTM) networks to adaptively optimize its parameters, thereby achieving faster convergence and improved surveillance coverage in urban transport networks. Second, we employ quantum LSTM (QLSTM) networks to predict traffic flow from data collected by the optimally placed sensors. The QLSTM model reduces the number of learnable parameters while maintaining the expressive capability of classical LSTM architectures. This semantic information is then leveraged to manage traffic flow more effectively, predicting traffic patterns and supporting proactive decision-making. Experimental results demonstrate that the LSTM-guided QAOA outperforms conventional optimization methods such as stochastic gradient descent, achieving faster and more reliable convergence. Likewise, the QLSTM model attains superior predictive accuracy, as evidenced by significant improvements in the explained variance score and root mean squared error. Collectively, these advancements represent a substantial step forward in intelligent traffic management and highlight the practical potential of HQC machine intelligence in IoIV systems.
A key component of the comprehending genetic variants and their biological effects is the rapid identification of protein mutations. For mutation classification, a novel machine learning framework is used in this study that combines quantum and classical deep learning techniques. Protein sequence data from the National Centre for Biotechnology Information (NCBI) is used and undergo preprocessing, artificial mutation simulation, and one-hot encoding to convert sequences into numerical representations. Feature selection is performed using Ant Colony Optimization (ACO) to reduce dimensionality and enhance computational efficiency. The classification process uses a Quantum Long Short-Term Memory (QLSTM) network with linear, variational quantum circuit (VQC), fully connected, dropout, and output layers. The hybrid model successfully separates mutant and non-mutated protein sequences and the hybrid methodology's overall accuracy was 95%, according to experimental assessment, providing a viable approach for bioinformatics applications that make use of quantum computing.