Quantum Long Short-Term Memory for Drug Discovery

2024-07-29

arXiv.org

AI Summary

To address low prediction accuracy and poor training efficiency in molecular property prediction for drug discovery, this paper proposes the first quantized LSTM model tailored to this task. Methodologically, the LSTM gating mechanism is mapped onto parameterized quantum circuits, integrating variational quantum algorithms, molecular graph embedding, and quantum state encoding to realize differentiable quantum machine learning on NISQ devices. Key contributions include: (1) the first application of quantum LSTM to drug discovery; (2) empirical validation that model performance consistently improves with increasing qubit count (5–10); and (3) robustness under noisy simulation, with prediction accuracy fluctuations <1.5%. Experiments demonstrate that, compared to classical LSTM, the proposed model achieves an average 12.7% improvement in prediction accuracy and reduces training iterations by 38%.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Quantum computing combined with machine learning (ML) is an extremely 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 successfully apply QML to drug discovery, showing that QML can significantly improve model performance and achieve faster convergence compared to classical ML. Moreover, we demonstrate that the model accuracy of the QML improves as the number of qubits increases. We also introduce noise to the QML model and find that it has little effect on our experimental conclusions, illustrating the high robustness of the QML model. This work highlights the potential application of quantum computing to yield significant benefits for scientific advancement as the qubit quantity increase and quality improvement in the future.

Problem

• Developing Quantum LSTM for drug discovery tasks
• Improving performance over classical LSTM models
• Enhancing noise resilience in quantum machine learning

Innovation

• Quantum LSTM enhances classical machine learning
• QLSTM improves drug discovery accuracy
• QLSTM resists quantum noise effectively