(CA) Scan-Parallel Variational Inference for event-driven Latent Jump Models

新的潜在跳跃模型为时间序列分析提供更快的训练速度

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-12 04:00

研究人员推出了一种新的时间序列分析框架——潜在跳跃模型（LJMs）。与依赖平滑ODE或SDE动力学的传统模型不同，LJMs包含由泊松驱动的离散事件，这些事件会导致瞬时状态转移。该模型具有状态依赖的强度和标记分布，定义了一个时间齐次的马尔可夫过程。为了训练，开发了一种新颖的事件增强路径空间证据下界（ELBO）和扫描并行变分推断（SPVI），通过在状态复现评估之前对潜在事件配置进行采样，从而实现更快、更具可扩展性的训练。 AI

影响引入了一种新颖的时间序列分析框架，其训练速度显著快于现有方法。

排序理由该集群包含一篇详细介绍新统计建模技术的 istatistical research paper。 [lever_c_demoted from research: ic=1 ai=0.7]

在 arXiv stat.ML 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍新统计建模技术的 istatistical research paper。 [lever_c_demoted from research: ic=1 ai=0.7]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv stat.ML TIER_1 (CA) · Vansh Bansal, Min Chen, James G. Scott · 2026-10-12 04:00

Scan-Parallel Variational Inference for Event-Driven Latent Jump Models

arXiv:2610.12986v1 Announce Type: new Abstract: Continuous-time latent state-space models are widely used for irregularly observed time series, typically representing uncertainty through a random initial state followed by smooth ordinary differential equation (ODE) dynamics, or B…

本文如何被排名

报道来源 [1]

Scan-Parallel Variational Inference for Event-Driven Latent Jump Models

相关实体

相关话题