Hybrid Classical-Quantum Variational Autoencoder for Neural Topic Modeling

用于神经主题建模的混合经典-量子变分自编码器

Abstract: Neural topic models enable scalable semantic discovery, but their integration with quantum hardware remains largely unexplored. We present a proof-of-concept hybrid classical-quantum variational autoencoder (VAE) for topic modeling, embedding parameterized quantum circuits within the VAE inference network while retaining a classical topic-word decoder.

摘要： 神经主题模型实现了可扩展的语义发现，但其与量子硬件的结合仍处于探索阶段。我们提出了一种用于主题建模的混合经典-量子变分自编码器（VAE）概念验证模型，该模型将参数化量子电路嵌入到 VAE 推理网络中，同时保留了经典的“主题-词”解码器。

To address the resource constraints of quantum hardware, we propose a modified Gaussian Softmax posterior that decouples latent space dimensionality from the number of topics to be extracted, enabling the model to operate with a low-resource 10-qubit quantum device.

为了解决量子硬件的资源限制，我们提出了一种改进的高斯 Softmax 后验分布，它将潜在空间的维度与待提取的主题数量解耦，使模型能够在资源受限的 10 量子比特设备上运行。

On the AgNews dataset, the hybrid VAE outperforms state-of-the-art neural topic models (NTMs), reaching a $C_v$ coherence score of 0.71 and an NPMI score of 0.20 while preserving high topic diversity. For comparison, we also construct a fully classical variant, which also outperforms state-of-the-art models on AgNews and exhibits clear class separation in the latent space.

在 AgNews 数据集上，该混合 VAE 的表现优于当前最先进的神经主题模型（NTMs），其 $C_v$ 一致性得分达到 0.71，NPMI 得分达到 0.20，同时保持了极高的主题多样性。作为对比，我们还构建了一个全经典变体，它在 AgNews 上同样优于现有最先进模型，并在潜在空间中展现出清晰的类别区分度。

These results demonstrate that hybrid VAEs are computationally viable even on NISQ-era devices and represent a promising direction for quantum-enhanced topic modeling.

这些结果表明，混合 VAE 即使在 NISQ（含噪声中等规模量子）时代的设备上也是计算可行的，并为量子增强主题建模提供了一个有前景的发展方向。