The (variational) graph auto-encoder is extensively employed for learning
representations of graph-structured data. However, the formation of real-world
graphs is a complex and heterogeneous process influenced by latent factors.
Existing encoders are fundamentally holistic, neglecting the entanglement of
latent factors. This not only makes graph analysis tasks less effective but
also makes it harder to understand and explain the representations. Learning
disentangled graph representations with (variational) graph auto-encoder poses
significant challenges, and remains largely unexplored in the existing
literature. In this article, we introduce the Disentangled Graph Auto-Encoder
(DGA) and Disentangled Variational Graph Auto-Encoder (DVGA), approaches that
leverage generative models to learn disentangled representations. Specifically,
we first design a disentangled graph convolutional network with multi-channel
message-passing layers, as the encoder aggregating information related to each
disentangled latent factor. Subsequently, a component-wise flow is applied to
each channel to enhance the expressive capabilities of disentangled variational
graph auto-encoder. Additionally, we design a factor-wise decoder, considering
the characteristics of disentangled representations. In order to further
enhance the independence among representations, we introduce independence
constraints on mapping channels for different latent factors. Empirical
experiments on both synthetic and real-world datasets show the superiority of
our proposed method compared to several state-of-the-art baselines.

我们介绍了 “解缠图自编码器” 和 “解缠变分图自编码器”（DGA 和 DVGA）的方法，其利用生成模型学习解缠表示，通过设计多通道信息传递层的解缠图卷积网络作为编码器，以及引入在不同潜在因素上的独立性约束，我们的方法在合成和真实数据集上实验表明其优越性。

利用解缠结构图自编码器学习网络表示

Learning Network Representations with Disentangled Graph Auto-Encoder

Recent empirical studies on domain generalization (DG) have shown that DG
algorithms that perform well on some distribution shifts fail on others, and no
state-of-the-art DG algorithm performs consistently well on all shifts.
Moreover, real-world data often has multiple distribution shifts over different
attributes; hence we introduce multi-attribute distribution shift datasets and
find that the accuracy of existing DG algorithms falls even further. To explain
these results, we provide a formal characterization of generalization under
multi-attribute shifts using a canonical causal graph. Based on the
relationship between spurious attributes and the classification label, we
obtain realizations of the canonical causal graph that characterize common
distribution shifts and show that each shift entails different independence
constraints over observed variables. As a result, we prove that any algorithm
based on a single, fixed constraint cannot work well across all shifts,
providing theoretical evidence for mixed empirical results on DG algorithms.
Based on this insight, we develop Causally Adaptive Constraint Minimization
(CACM), an algorithm that uses knowledge about the data-generating process to
adaptively identify and apply the correct independence constraints for
regularization. Results on fully synthetic, MNIST, small NORB, and Waterbirds
datasets, covering binary and multi-valued attributes and labels, show that
adaptive dataset-dependent constraints lead to the highest accuracy on unseen
domains whereas incorrect constraints fail to do so. Our results demonstrate
the importance of modeling the causal relationships inherent in the
data-generating process.

本论文提出了一种适应性统计算法 CACM，通过建立数据生成过程中的因果关系来自适应地识别并应用正确的独立性约束，从而提高对不可见域的准确性，证明了现有单一、固定约束算法无法在所有转移中有效而导致了混合实证结果。

建模数据生成过程是实现超出分布泛化的必要条件

Modeling the Data-Generating Process is Necessary for Out-of-Distribution Generalization

We consider the problem of discovering causal relations from independence
constraints selection bias in addition to confounding is present. While the
seminal FCI algorithm is sound and complete in this setup, no criterion for the
causal interpretation of its output under selection bias is presently known. We
focus instead on local patterns of independence relations, where we find no
sound method for only three variable that can include background knowledge.
Y-Structure patterns are shown to be sound in predicting causal relations from
data under selection bias, where cycles may be present. We introduce a
finite-sample scoring rule for Y-Structures that is shown to successfully
predict causal relations in simulation experiments that include selection
mechanisms. On real-world microarray data, we show that a Y-Structure variant
performs well across different datasets, potentially circumventing spurious
correlations due to selection bias.

本研究探讨如何处理因选择偏差而导致的独立性约束偏差问题，并通过独立关系的局部模式，提出了一种基于 Y-Structure 的有限样本计分规则，成功地预测了选择机制下的因果关系，并在微阵列数据上表现出色。