Research map / Graph learning / Graph algorithms and clustering
Matrix Completion with Hierarchical Graph Side Information
Adel Elmahdy, Junhyung Ahn, Changho Suh, Soheil Mohajer
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add local refinement of rating vectors in addition to groups (flag=1 vs flag=0) (ablation) | hierarchical matrix completion algorithm · synthetic, n=3000, m=1000 | success rate | Improved |
| increase graph quality alpha tilde | two refinement strategies · synthetic, n=3000, m=1000 | gap between refinement strategies | Improved |
| increase n and m with ratio n/m=3 | proposed algorithm · synthetic, perfect clustering and grouping-limited regimes | success rate phase transition at p* | Improved |
| use proposed algorithm instead of baselines (User/Item Avg, k-NN, TrustSVD, Biased MF, SoReg) | proposed algorithm · Poliblog semi-real, sigma^2=0.5, varying p | RMSE | Improved |
| use proposed algorithm instead of baselines under varying noise | proposed algorithm · Poliblog semi-real, p=0.08, varying 1/sigma^2 | RMSE | Improved |
| compare runtime of proposed algorithm to User k-NN | proposed algorithm · Poliblog, p=0.1 | runtime (sec): 0.299 -> 0.055 | Improved |
| compare runtime of proposed algorithm to Biased MF | proposed algorithm · Poliblog, p=0.1 | runtime (sec): 0.266 -> 0.055 | Improved |
| compare runtime of proposed algorithm to TrustSVD | proposed algorithm · Poliblog, p=0.1 | runtime (sec): 0.482 -> 0.055 | Improved |
| compare runtime of proposed algorithm to User Average | proposed algorithm · Poliblog, p=0.1 | runtime (sec): 0.021 -> 0.055 | Hurt |
| compare runtime of proposed algorithm to Item Average | proposed algorithm · Poliblog, p=0.1 | runtime (sec): 0.025 -> 0.055 | Hurt |
Closest papers by meaning
- Mixture Matrix Completion (NeurIPS 2018 · similarity 0.89)
- Scalable Completion of Nonnegative Matrix with Separable Structure (AAAI 2016 · similarity 0.88)
- High Rank Matrix Completion With Side Information (AAAI 2018 · similarity 0.88)
- A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering (ICML 2024 · similarity 0.87)
- Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability (ICML 2025 · similarity 0.86)
- Online Matrix Completion: A Collaborative Approach with Hott Items (ICML 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
