---
id: geometric-deep-learning
title: "Geometric Deep Learning: Group Equivariance and Symmetry"
schema_type: article
category: ai
language: en
confidence: medium
last_verified: "2026-05-28"
created_date: "2026-05-24"
generation_method: ai_structured
ai_models:
  - claude-4.5-sonnet
derived_from_human_seed: true
conflict_of_interest: none_declared
is_live_document: false
data_period: static
completeness: 0.85
atomic_facts:
  - id: af-ai-geometric-deep-learning-1
    statement: >-
      The geometric deep learning survey frames the field around deep learning on structured domains
      such as graphs, groups, and manifolds.
    source_title: "Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges"
    source_url: https://arxiv.org/abs/2104.13478
    confidence: medium
  - id: af-ai-geometric-deep-learning-2
    statement: >-
      Graph Convolutional Networks apply neural network layers to graph-structured data for
      semi-supervised classification.
    source_title: Semi-Supervised Classification with Graph Convolutional Networks
    source_url: https://arxiv.org/abs/1609.02907
    confidence: medium
  - id: af-ai-geometric-deep-learning-3
    statement: PointNet directly processes point sets for 3D classification and segmentation tasks.
    source_title: "PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation"
    source_url: https://arxiv.org/abs/1612.00593
    confidence: medium
primary_sources:
  - id: ps-ai-geometric-deep-learning-1
    title: "Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges"
    type: academic_paper
    year: 2021
    institution: arXiv
    url: https://arxiv.org/abs/2104.13478
  - id: ps-ai-geometric-deep-learning-2
    title: Semi-Supervised Classification with Graph Convolutional Networks
    type: academic_paper
    year: 2016
    institution: arXiv
    url: https://arxiv.org/abs/1609.02907
  - id: ps-ai-geometric-deep-learning-3
    title: "PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation"
    type: academic_paper
    year: 2016
    institution: arXiv
    url: https://arxiv.org/abs/1612.00593
known_gaps:
  - Efficient equivariant networks at scale
  - Equivariant architectures for video and temporal data
disputed_statements: []
secondary_sources: []
updated: "2026-05-28"
---
## TL;DR
Geometric Deep Learning: Group Equivariance and Symmetry: Geometric deep learning extends neural networks to data with structure such as graphs, manifolds, sets, meshes, and point clouds.

## Core Explanation
The field studies how symmetry, invariance, equivariance, and domain geometry shape model architectures. Graph convolutional networks and point-cloud networks are practical examples.

## Further Reading

- [Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges](https://arxiv.org/abs/2104.13478)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://arxiv.org/abs/1609.02907)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://arxiv.org/abs/1612.00593)