Bayesian Self-Supervised Learning Using Local and Global Graph Information
Bayesian Self-Supervised Learning Using Local and Global Graph Information
复制标题
DOI：
10.1109/camsap58249.2023.10403487
复制
发表时间：
2023-12
期刊：
影响因子：
--
通讯作者：
Konstantinos D. Polyzos;A. Sadeghi;G. Giannakis
中科院分区：
文献类型：
--
作者：
Konstantinos D. Polyzos;A. Sadeghi;G. Giannakis
图形引导学习在一系列网络科学应用中的影响是有据可查的。典型的图引导学习任务处理图上的半监督学习，其中的目标是通过利用少量的节点观测以及底层的图结构来预测未观察到的节点的节点值或标签。这在隐私限制下或通常在获取节点观测引起高成本的情况下尤其具有挑战性。在此背景下，本工作提出了一种贝叶斯图驱动的自监督学习(SELF-SL)方法：(I)学习从更容易求解的辅助任务产生的强大的节点嵌入，这些辅助任务将局部连通性信息映射到全局连通性信息；以及(Ii)采用具有自适应权重的高斯过程(EGP)集成作为节点嵌入的在线处理。与大多数现有的确定性方法不同，新方法提供了对未观察到的节点值的准确估计以及不确定性量化，这在安全关键应用中尤其重要。在合成和真实图形数据集上的数值测试显示了新的基于EGP的SELF-SL方法的优点。
Graph-guided learning has well-documented impact in a gamut of network science applications. A prototypical graph-guided learning task deals with semi-supervised learning over graphs, where the goal is to predict the nodal values or labels of unobserved nodes, by leveraging a few nodal observations along with the underlying graph structure. This is particularly challenging under privacy constraints or generally when acquiring nodal observations incurs high cost. In this context, the present work puts forth a Bayesian graph-driven self-supervised learning (Self-SL) approach that: (i) learns powerful nodal embeddings emanating from easier to solve auxiliary tasks that map local to global connectivity information; and, (ii) adopts an ensemble of Gaussian processes (EGPs) with adaptive weights as nodal embeddings are processed online. Unlike most existing deterministic approaches, the novel approach offers accurate estimates of the unobserved nodal values along with uncertainty quantification that is important especially in safety critical applications. Numerical tests on synthetic and real graph datasets showcase merits of the novel EGP-based Self-SL method.