mlsanityData to 14 Oct 2026

Adversarial Erasing with Pruned Elements: Towards Better Graph Lottery Ticket

arXiv 2308.02916 ↗First posted 5 Aug 2023cs.LG cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Graph Lottery Ticket (GLT), a combination of core subgraph and sparse subnetwork, has been proposed to mitigate the computational cost of deep Graph Neural Networks (GNNs) on large input graphs while preserving original performance. However, the winning GLTs in exisiting studies are obtained by applying iterative magnitude-based pruning (IMP) without re-evaluating and re-considering the pruned information, which disregards the dynamic changes in the significance of edges/weights during graph/model structure pruning, and thus limits the appeal of the winning tickets. In this paper, we formulate a conjecture, i.e., existing overlooked valuable information in the pruned graph connections and model parameters which can be re-grouped into GLT to enhance the final performance. Specifically, we propose an adversarial complementary erasing (ACE) framework to explore the valuable information from the pruned components, thereby developing a more powerful GLT, referred to as the ACE-GLT. The main idea is to mine valuable information from pruned edges/weights after each round of IMP, and employ the ACE technique to refine the GLT processing. Finally, experimental results demonstrate that our ACE-GLT outperforms existing methods for searching GLT in diverse tasks. Our code will be made publicly available.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 28 of its 37 references that are in the map. First posted on arXiv 5 Aug 2023.

Builds on

28 of 37 references.

References · 28

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2017Inductive Representation Learning on Large Graphscited by 6,320
  4. 2018How Powerful are Graph Neural Networks?cited by 4,393
  5. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  6. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  7. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
  8. 2020Simple and Deep Graph Convolutional Networkscited by 860
  9. 2018Hierarchical Graph Representation Learning with Differentiable Poolingcited by 787
  10. 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 680
  11. 2019Self-Attention Graph Poolingcited by 421
  12. 2020DeeperGCN: All You Need to Train Deeper GCNscited by 250
16 more
  1. 2020Comparing Rewinding and Fine-tuning in Neural Network Pruningcited by 240
  2. 2019Drawing Early-Bird Tickets: Towards More Efficient Training of Deep Networkscited by 160
  3. 2021Graph Condensation for Graph Neural Networkscited by 150
  4. 2020Learning to Drop: Robust Graph Neural Network via Topological Denoisingcited by 127
  5. 2021A Unified Lottery Ticket Hypothesis for Graph Neural Networkscited by 118
  6. 2020Neural Pruning via Growing Regularizationcited by 97
  7. 2020Degree-Quant: Quantization-Aware Training for Graph Neural Networkscited by 96
  8. 2019GraphZoom: A multi-level spectral approach for accurate and scalable graph embeddingcited by 58
  9. 2020Scattering GCN: Overcoming Oversmoothness in Graph Convolutional Networkscited by 57
  10. 2018Spectrally approximating large graphs with smaller graphscited by 56
  11. 2020The Lottery Ticket Hypothesis for Object Recognitioncited by 34
  12. 2021Early-Bird GCNs: Graph-Network Co-Optimization Towards More Efficient GCN Training and Inference via Drawing Early-Bird Lottery Ticketscited by 27
  13. 2021Distribution Knowledge Embedding for Graph Poolingcited by 19
  14. 2022Comprehensive Graph Gradual Pruning for Sparse Training in Graph Neural Networkscited by 18
  15. 2022You Can Have Better Graph Neural Networks by Not Training Weights at All: Finding Untrained GNNs Ticketscited by 14
  16. 2022Learning to Win Lottery Tickets in BERT Transfer via Task-agnostic Mask Trainingcited by 10

Built on by

Cited by year

Citing papers per year: 2024: 5.

5 papers.

  1. 2024EXGC: Bridging Efficiency and Explainability in Graph Condensationcited by 20
  2. 2024Graph Sparsification via Mixture of Graphscited by 17
  3. 2024Two Heads Are Better Than One: Boosting Graph Sparse Training via Semantic and Topological Awarenesscited by 10
  4. 2024Unveiling Global Interactive Patterns across Graphs: Towards Interpretable Graph Neural Networkscited by 8
  5. 2024Fast Track to Winning Tickets: Repowering One-Shot Pruning for Graph Neural Networkscited by 0