Euclidean geometry meets graph, a geometric deep learning perspective
Graph neural networks (GNN) have been an active area of machine learning research to tackle various problems in graph data. Sometimes nodes and edges in a graph can have spatial features, such as 3D coordinates of nodes and directions along edges. How do we reason over the topology of graphs while considering those geometric features? In this post, we discuss a ICLR 2021 paper "Learning from Protein Structure with Geometric Vector Perceptrons (GVP)". We take a geometric deep learning perspective to explain the intuition and advantage of GVP-GNN, the novel GNN architecture described in the paper, to reason over geometric graphs. We inspire an idea along physics-oriented interpretation. We also highlight under-appreciated applications of geometric graph learning in solving problems from multiple domains, including biochemistry, geography, and network analysis.
Code (0)
등록된 구현이 없습니다.
Tasks
Deep LearningGraph LearningSimilar Papers 제목 키워드 기반
Beyond Euclid: An Illustrated Guide to Modern Machine Learning with Geometric, Topological, and Algebraic Structures
The enduring legacy of Euclidean geometry underpins classical machine learning, which, for decades, has been primarily developed for data lying in Euclidean space. Yet, modern machine learning increasingly encounters ric…
Exploring Data Geometry for Continual Learning
Continual learning aims to efficiently learn from a non-stationary stream of data while avoiding forgetting the knowledge of old data. In many practical applications, data complies with non-Euclidean geometry. As such, t…
Continual LearningGeometry Contrastive Learning on Heterogeneous Graphs
Self-supervised learning (especially contrastive learning) methods on heterogeneous graphs can effectively get rid of the dependence on supervisory data. Meanwhile, most existing representation learning methods embed the…
Contrastive LearningNode ClassificationNode ClusteringRepresentation Learning+1Learning Geometry-Dependent and Physics-Based Inverse Image Reconstruction
Deep neural networks have shown great potential in image reconstruction problems in Euclidean space. However, many reconstruction problems involve imaging physics that are dependent on the underlying non-Euclidean geomet…
Image ReconstructionNode-Specific Space Selection via Localized Geometric Hyperbolicity in Graph Neural Networks
Many graph neural networks have been developed to learn graph representations in either Euclidean or hyperbolic space, with all nodes' representations embedded in a single space. However, a graph can have hyperbolic and …
Graph Neural NetworkLink PredictionNode Classification