EUVIMEDEUVIMED entdecken
Crossref

Memory Is No Longer a Bottleneck: Memory-Efficient Graph Filtering for Scalable Collaborative Filtering

Jin-Duk Park, Won-Yong Shin

IEEE Transactions on Knowledge and Data Engineering · 2026-08 · 38 (8) · 5269-5281

DOI: 10.1109/tkde.2026.3704986 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Knowledge and Data Engineering, 2026-08. Verfasst von Jin-Duk Park und 1 weiteren Autor:innen. Die Arbeit nennt 66 Literaturangaben. Gefördert von: National Research Foundation of Korea, Korea Government, SMEs Technology Innovation Development Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Graph convolutional networks (GCNs) have demonstrated significant success in capturing complex user–item relationships for collaborative filtering (CF). However, due to their reliance on extensive model training, training-free graph filtering (GF)-based CF methods have emerged as a promising alternative, offering computational efficiency by smoothing graph signals via matrix operations. In particular, polynomial GF-based approaches demonstrate improved accuracy through their ability to design more expressive and flexible filtering functions. Despite these advantages, existing GF methods suffer from a critical memory bottleneck: they necessitate storing the full item similarity graph, incurring prohibitive memory costs for large-scale datasets, which limits their practical applicability. To tackle this challenge, we propose ${\sf Mem\! -\! GF}$ Mem - GF ( Mem ory-efficient GF ), a new GF-based CF method that departs from conventional designs by principally leveraging the structure of Krylov subspaces as a core mechanism for approximating polynomial graph filters without explicitly storing the item similarity graph. We theoretically analyze the minimum Krylov subspace size that guarantees lossless approximation. Through extensive experiments, we demonstrate that ${\sf Mem\! -\! GF}$ Mem - GF achieves up to 5.74 × lower memory usage and 4.38 × speedup in runtime, while consistently exceeding the recommendation accuracy of state-of-the-art GF and GCN-based methods. ${\sf Mem\! -\! GF}$ Mem - GF robustly scales to datasets with tens of millions of interactions, establishing itself as a practically viable and theoretically grounded solution for efficient CF.

Förderung

  • National Research Foundation of Korea
  • Korea Government
  • SMEs Technology Innovation Development Program
  • Technology Innovation and Promotion Agency
  • Ministry of SMEs and Startups

Zitieren

APA

Park, J., & Shin, W. (2026). Memory Is No Longer a Bottleneck: Memory-Efficient Graph Filtering for Scalable Collaborative Filtering. IEEE Transactions on Knowledge and Data Engineering, 38(8), 5269-5281. https://doi.org/10.1109/tkde.2026.3704986

Vancouver

Park J, Shin W. Memory Is No Longer a Bottleneck: Memory-Efficient Graph Filtering for Scalable Collaborative Filtering. IEEE Transactions on Knowledge and Data Engineering. 2026;38(8):5269-5281. doi:10.1109/tkde.2026.3704986

BibTeX
@article{Park2026,
  author = {Park, Jin-Duk and Shin, Won-Yong},
  title = {Memory Is No Longer a Bottleneck: Memory-Efficient Graph Filtering for Scalable Collaborative Filtering},
  journal = {IEEE Transactions on Knowledge and Data Engineering},
  year = {2026},
  volume = {38},
  number = {8},
  pages = {5269--5281},
  doi = {10.1109/tkde.2026.3704986}
}
Kostenlos starten →