Skip to main content
QUICK REVIEW

[论文解读] Neural Backward Filtering Forward Guiding

Gefan Yang, Frank van der Meulen|arXiv (Cornell University)|Jan 30, 2026
Morphological variations and asymmetry被引用 0
一句话总结

NBFFG 构建了一个变分框架,将一个解析可处理的向后引导与神经残差结合,用以在离散与连续动力学下对树上的平滑后验进行推断,支持路径级子采样与摊销学习。它在线性、多峰和高维的系统发育任务中显示出改进的性能。

ABSTRACT

Inference in non-linear continuous stochastic processes on trees is challenging, particularly when observations are sparse (leaf-only) and the topology is complex. Exact smoothing via Doob's $h$-transform is intractable for general non-linear dynamics, while particle-based methods degrade in high dimensions. We propose Neural Backward Filtering Forward Guiding (NBFFG), a unified framework for both discrete transitions and continuous diffusions. Our method constructs a variational posterior by leveraging an auxiliary linear-Gaussian process. This auxiliary process yields a closed-form backward filter that serves as a ``guide'', steering the generative path toward high-likelihood regions. We then learn a neural residual--parameterized as a normalizing flow or a controlled SDE--to capture the non-linear discrepancies. This formulation allows for an unbiased path-wise subsampling scheme, reducing the training complexity from tree-size dependent to path-length dependent. Empirical results show that NBFFG outperforms baselines on synthetic benchmarks, and we demonstrate the method on a high-dimensional inference task in phylogenetic analysis with reconstruction of ancestral butterfly wing shapes.

研究动机与目标

  • 在叶节点观测且拓扑复杂的非线性、树结构随机过程上动机化平滑。
  • 开发一个统一框架,将解析的向后引导与神经纠正融合,适用于离散与连续动力学。
  • 通过路径级子采样与树内摊销学习降低训练复杂度,以扩展到大规模树。
  • 在线性基准、多模态非线性系统和高维系统发育任务上展示鲁棒性与准确性。

提出的方法

  • 通过用可处理的辅助线性高斯过程替代不可处理的 h 函数来构建带引导的提案,从而获得闭式向后滤波。
  • 将变分后验定义为神经残差——对离散边用正则化流参数化,或对连续路径用神经性 SDE 参数化,以捕捉非线性差异。
  • 通过变分后验与真实后验之间的 KL 散度,以及叶观测似然性的项(ELBO 形式)来计算损失。
  • 使用两条推论推论推论来表达离散与连续动力学的损失,使得在引导提案与神经变换下的可处理计算成为可能。
  • 应用路径级子采样方案,以获得全树损失的无偏估计,将计算量从树大小降至路径长度。
Figure 1 : Validation on Linear Gaussian Benchmarks. We compare the converged training loss against the analytical RTS smoother baseline. (a) Topological Scalability: Relative error decreases as tree complexity ( $N_{\mathrm{depth}},N_{\mathrm{branch}}$ ) grows, showing that our path-wise amortizati
Figure 1 : Validation on Linear Gaussian Benchmarks. We compare the converged training loss against the analytical RTS smoother baseline. (a) Topological Scalability: Relative error decreases as tree complexity ( $N_{\mathrm{depth}},N_{\mathrm{branch}}$ ) grows, showing that our path-wise amortizati

实验结果

研究问题

  • RQ1混合引导提案框架是否能近似 Doob h-变换后验,用于离散与连续动力学下的树上平滑?
  • RQ2神经残差(正则化流或神经 SDE)在多大程度上能弥补线性高斯引导之外的漂移与非线性?
  • RQ3路径级子采样是否能在保持后验精度的同时实现无偏、可扩展的训练,并在树深度与维度上保持稳定?
  • RQ4带有在树内共享网络的摊销学习是否足以在复杂、异构的树拓扑之间实现泛化?
  • RQ5NBFFG在线性基准、多模态非线性系统与高维系统发育重建任务上的表现如何?

主要发现

  • NBFFG 框架通过将可处理的引导提案与神经残差结合,产生变分后验,使离散与连续树边的学习高效实现。
  • 路径级、无偏估计使每次迭代仅在单一根根到叶路径上进行训练,降低计算量从树大小到路径长度。
  • 在线性模型中,该方法与解析真值高度一致;在多模态后验情况下,当基线引导因模式崩溃而失败时,仍能恢复后验。
  • 实验结果表明对高维度具有鲁棒性,方法能有效处理扩散桥与非线性动力学,在合成与系统发育情境中均有效。
  • 带有对父状态、边持续时间和上下文的共享网络条件的摊销学习,使大规模和不规则树的推断具有可扩展性。
Figure 2 : Empirical distributions of 500 independent samples of the guided proposal (gray) and the learned variational posterior (orange) against the analytical ground truth (RTS, green contours) at different tree depths.
Figure 2 : Empirical distributions of 500 independent samples of the guided proposal (gray) and the learned variational posterior (orange) against the analytical ground truth (RTS, green contours) at different tree depths.

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。