NODAGS-Flow: Nonlinear Cyclic Causal Structure Learning
NODAGS-Flow: Nonlinear Cyclic Causal Structure Learning
复制标题
DOI：
10.48550/arxiv.2301.01849
复制
发表时间：
2023-01
期刊：
影响因子：
--
通讯作者：
Muralikrishnna G. Sethuraman;Romain Lopez;Ramkumar Veppathur Mohan;F. Fekri;Tommaso Biancalani;Jan-Christian Hutter
中科院分区：
文献类型：
--
作者：
Muralikrishnna G. Sethuraman;Romain Lopez;Ramkumar Veppathur Mohan;F. Fekri;Tommaso Biancalani;Jan-Christian Hutter
学习变量之间的因果关系是统计学中一个经过充分研究的问题，在科学中有许多重要的应用。然而，对现实世界系统进行建模仍然具有挑战性，因为大多数现有算法都假设底层因果图是非循环的。虽然这是开发有关因果推理和推理的理论发展的便捷框架，但在实际系统中可能会违反潜在的建模假设，因为反馈循环很常见（例如，在生物系统中）。尽管有一些方法搜索循环因果模型，但它们通常依赖于某种形式的线性，这也是有限制的，或者缺乏明确的潜在概率模型。在这项工作中，我们提出了一种新的框架，用于从介入数据中学习非线性循环因果图模型，称为 NODAGS-Flow。我们通过直接似然优化进行推理，采用残差归一化流的技术进行似然估计。通过合成实验和对单细胞高内涵扰动筛选数据的应用，我们的方法在结构恢复和预测性能方面与最先进的方法相比显示出显着的性能改进。
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.