来自今天的爱可可AI前沿推介

[LG] NODAGS-Flow: Nonlinear Cyclic Causal Structure Learning

M G. Sethuraman, R Lopez, R Mohan, F Fekri, T Biancalani, J Hütter
[Georgia Institute of Technology & Stanford University & Genentech]

NODAGS-Flow: 非线性循环因果结构学习

要点:

  1. NODAGS-Flow 是一种从干预数据学习非线性环形因果关系的新框架；
  2. NODAGS-Flow 基于将观察模型制定为离散动力学系统的稳态，并用归一化流进行可能性估计；
  3. 在合成实验和真实世界的基因表达数据中进行基因干预时，NODAGS-Flow 与最先进方法相比，在结构恢复和预测性能方面表现出显著的性能提升。

一句话总结:
NODAGS-Flow 是一种新的因果发现方法，能通过简单的优化框架学习变量间的非线性和环形关系，并在结构恢复和预测任务方面比最先进方法表现更好。

摘要：
学习变量间的因果关系是统计学中研究较多的问题，在科学中有很多重要应用。然而，对于真实世界的系统建模仍然具有挑战性，因为大多数现有算法假设基础因果图是无环的。虽然这是开发因果推理和推断的理论发展的方便框架，但是在真实系统中，因为反馈回路很常见(例如，在生物系统中)，所以基础建模假设可能会打破。尽管有几种方法搜索环形因果模型，但通常依赖于某种形式的线性关系，这也是有限的，或缺乏明确的基础概率模型。本文提出一种新框架，用于从干预数据学习非线性环形因果关系的图模型，即 NODAGS-Flow。通过直接似然优化执行推断，利用残差归一化流的似然估计技术。通过合成实验和应用于单细胞高内容干扰筛查数据，展示了该方法与最先进方法相比在结构发现和预测性能方面具有优越性。

Learning causal relationships between variables is a well-studied problem in statistics, with many important applications in science. However, modeling real-world systems remain challenging, as most existing algorithms assume that the underlying causal graph is acyclic. While this is a convenient framework for developing theoretical developments about causal reasoning and inference, the underlying modeling assumption is likely to be violated in real systems, because feedback loops are common (e.g., in biological systems). Although a few methods search for cyclic causal models, they usually rely on some form of linearity, which is also limiting, or lack a clear underlying probabilistic model. In this work, we propose a novel framework for learning nonlinear cyclic causal graphical models from interventional data, called NODAGS-Flow. We perform inference via direct likelihood optimization, employing techniques from residual normalizing flows for likelihood estimation. Through synthetic experiments and an application to single-cell high-content perturbation screening data, we show significant performance improvements with our approach compared to state-of-the-art methods with respect to structure recovery and predictive performance.

论文链接：https://arxiv.org/abs/2301.01849
图片
图片
图片
图片

内容中包含的图片若涉及版权问题，请及时与我们联系删除