改进神经马尔可夫逻辑网络，提升大图生成能力。
Parallel Noising in Neural Markov Logic Networks
- 用图神经网络增强势函数表达能力。
- 提出并行去噪算法，显著提升生成效果。
- 在小分子生成上媲美专用文本模型。
神经马尔可夫逻辑网络（NMLNs）是一种灵活的神经符号关系模型。先前研究表明，尽管NMLNs在小型关系结构生成中表现优异，但在大型结构上仍逊于基于扩散的图生成模型。本文从两方面强化NMLNs：(i) 利用图神经网络提升其势函数的表达能力；(ii) 提出一种受并行退火马尔可夫链蒙特卡洛方法启发的新训练与推理算法，命名为并行去噪。两项改进共同使NMLNs在图生成性能上达到与通用扩散模型相当的水平，并在小分子结构生成任务中匹配专门的基于文本的循环模型表现。
原文摘要 · Abstract (English)
Neural Markov Logic Networks (NMLNs) are a flexible neurosymbolic relational model. Previous work has shown that, although NMLNs achieve strong performance as generative models for small relational structures, they underperform diffusion-based generative graph models on larger structures. In this paper, we strengthen NMLNs along two main dimensions: (i) we increase the expressive capacity of their potential functions using graph neural networks, and (ii) we develop a new training and inference algorithm inspired by parallel-tempering Markov chain Monte Carlo methods, which we name parallel noising. Together, these enhancements enable NMLNs to attain strong performance in graph generation relative to general diffusion-based generative graph models. Furthermore, they allow NMLNs to match the performance of specialized text-based recurrent models when generating small molecular structures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。