Quantum physics: Long-term memory

Bao-Nhi Dang Tran, Muhammad Fahim, Adnan Ahmad Cheema +4 more

The quantification of physical activity energy expenditure (PAEE) offers significant benefits for healthcare monitoring and has the potential to promote healthy and active aging for elderly individuals. With recent advancements in quantum information and computation, quantum machine learning (QML) has emerged as a powerful tool capable of improving upon the measurement of PAEE. In this study, we propose a hybrid machine-learning model to predict PAEE. This model specifically leverages a classical long short-term memory (LSTM) model integrated with a variational quantum circuit (VQC). This model, which we refer to as the enhanced quantum long short-term memory linear (eQLSTML) model, was subsequently trained and tested using the publicly available GOTOV Human Physical Activity and Energy Expenditure Dataset for Older Individuals. Upon performance comparisons between the classical LSTM and proposed eQLSTML models, our findings suggest that the eQLSTML modeling approach demonstrates superior performance compared to classical machine learning methods, thereby holding a promise for personalized healthcare monitoring and promoting healthy aging in the older population.

Ran Zhao, Yaroslav O. Dudin, S. D. Jenkins +4 more

Bao-Nhi Dang Tran, Muhammad Fahim, B. D. E. McNiven +3 more

Physical activity energy expenditure (PAEE) offers significant benefits for general healthcare monitoring and has the potential to promote healthy and active aging for elderly individuals. With recent advancements in quantum information and computation, quantum machine learning (QML) has emerged as a tool capable of improving upon the measurement of PAEE. In this paper, we propose a hybrid QML model to predict PAEE which consists of a classical long short-term memory (LSTM) model integrated with a variational quantum circuit (VQC). This model, which we refer to as the enhanced quantum long short-term memory linear (eQLSTML), was subsequently trained and tested using the publicly available GOTOV Human Physical Activity and Energy Expenditure Dataset for Older Individuals. In particular, we study the proposed eQLSTML model with different gate choices in the quantum circuit along with various embedding and layering techniques. Our results indicate our model to be superior in both performance comparisons and prediction when compared to traditional machine learning methods currently employed. Our findings indicate that combining QML approaches with wearable IoT healthcare devices provides a new avenue for personalized healthcare monitoring and an effective method for promoting healthy aging.

Ramón Alonso‐Sanz

Sarvapriya M. Tripathi, Himanshu Upadhyay, Jayesh Soni

With the advent of quantum computing, researchers have explored the applicability and any potential advantages of quantum algorithms. This study investigates the application of Quantum Machine Learning (QML) models for regression tasks. Utilizing two distinct CERN datasets (Dielectron events and Proton collision), we investigate prediction accuracy using two QML algorithms, namely Quantum Neural Network (QNN) and Quantum Long Short-Term Memory (QLSTM). We also discuss the comparative analysis and computational efficiency of QNN and QLSTM compared to classical regression methods. The results show that while the two QML models gave comparable accuracy scores to some of the classical models, the best scores were still achieved by some of the more advanced classical algorithms such as CatBoost. Further analysis of the QNN and QLSTM algorithms using multiple ansätz designs showed that increased circuit complexity did not yield substantial improvements in prediction accuracy. These findings suggest that QML models, especially QLSTM with simpler ansätz designs, offer a promising approach for modeling high-energy physics data, and highlight the importance of balancing circuit complexity with performance. The study also underscores the need for further evaluation of these algorithms on quantum hardware to better understand real-world applicability.

Christoph Simon, M. Afzelius, J. Appel +21 more