arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.18997quant-ph
收藏

应用于药物毒性预测的量子到经典神经网络迁移学习

Quantum to Classical Neural Network Transfer Learning Applied to Drug Toxicity Prediction

Anthony M. Smaldone, Victor S. Batista

AI总结:

本文提出量子-经典混合神经网络,以 $\mathcal{O}(n^2)$ 量子矩阵乘法和 Hadamard 测试预测药物毒性,并将权重迁移至经典模型继续训练。

摘要:

毒性是阻碍过多药物用于可能挽救生命的应用的一个障碍。深度学习为寻找理想候选药物提供了一种有前景的解决方案;然而,化学空间的广阔性以及底层的 $\mathcal{O}(n^3)$ 矩阵乘法意味着这些工作很快会产生高昂的计算需求。为弥补这一点,我们提出了一种用于预测药物毒性的量子-经典混合神经网络,该网络采用一种模拟经典神经行为的量子电路设计,通过以 $\mathcal{O}(n^2)$ 复杂度显式计算矩阵乘积来实现。我们利用 Hadamard 测试进行高效内积估计,而不是传统使用的 swap 测试,从而将量子比特数量减少一半,并消除了对量子相位估计的需求。直接以量子力学方式计算矩阵乘积,使可学习权重能够从量子设备迁移到经典设备以进行进一步训练。我们将该框架应用于 Tox21 数据集,并表明其取得了与该模型完全经典的 $\mathcal{O}(n^3)$ 对应模型相当的预测准确率。此外,我们证明模型在迁移到完全经典架构后能够持续学习而不中断。我们相信,将复杂度降低的量子优势与无噪声计算的经典优势相结合,将为更具可扩展性的机器学习模型铺平道路。

英文摘要:

Toxicity is a roadblock that prevents an inordinate number of drugs from being used in potentially life-saving applications. Deep learning provides a promising solution to finding ideal drug candidates; however, the vastness of chemical space coupled with the underlying $\mathcal{O}(n^3)$ matrix multiplication means these efforts quickly become computationally demanding. To remedy this, we present a hybrid quantum-classical neural network for predicting drug toxicity, utilizing a quantum circuit design that mimics classical neural behavior by explicitly calculating matrix products with complexity $\mathcal{O}(n^2)$. Leveraging the Hadamard test for efficient inner product estimation rather than the conventionally used swap test, we reduce the number qubits by half and remove the need for quantum phase estimation. Directly computing matrix products quantum mechanically allows for learnable weights to be transferred from a quantum to a classical device for further training. We apply our framework to the Tox21 dataset and show that it achieves commensurate predictive accuracy to the model's fully classical $\mathcal{O}(n^3)$ analog. Additionally, we demonstrate the model continues to learn, without disruption, once transferred to a fully classical architecture. We believe combining the quantum advantage of reduced complexity and the classical advantage of noise-free calculation will pave the way to more scalable machine learning models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑