提出并行采样方法，让事件序列模型高效生成多步未来事件。
Speculative Sampling for Parametric Temporal Point Processes
- 基于拒绝采样，实现现有模型的并行多步采样
- 真实数据集上显著提升采样速度，无需修改模型或重新训练
- 适合需要高效生成大量事件序列的工业应用
时间点过程是建模事件序列的强大生成模型，能捕捉时间序列中的复杂依赖关系。通常采用自回归模型学习下一个事件的概率分布，导致采样必须串行进行，效率受限。本文提出一种基于拒绝采样的新算法，可在不改变模型架构或重新训练的前提下，对现有时间点过程模型实现多步未来事件的精确并行采样。该方法不仅具有理论保证，还在真实数据集上展现出显著的采样加速效果，为大规模时间点过程应用在表达能力与生成效率之间架起桥梁。
原文摘要 · Abstract (English)
Temporal point processes are powerful generative models for event sequences that capture complex dependencies in time-series data. They are commonly specified using autoregressive models that learn the distribution of the next event from the previous events. This makes sampling inherently sequential, limiting efficiency. In this paper, we propose a novel algorithm based on rejection sampling that enables exact sampling of multiple future values from existing TPP models, in parallel, and without requiring any architectural changes or retraining. Besides theoretical guarantees, our method demonstrates empirical speedups on real-world datasets, bridging the gap between expressive modeling and efficient parallel generation for large-scale TPP applications.
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。