Research map / Probabilistic and scientific modeling / Molecules and proteins
Nonparametric Sparse Tensor Factorization with Hierarchical Gamma Processes
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use sparse tensor modeling instead of dense factorization | Entry-value prediction · Alog, MovieLens, and SG | MSE and MAE | Helped |
| Use hierarchical Gamma processes with flexible location/sociability factor counts rather than NEST's restricted factor configurations | Ours versus NEST · Alog, MovieLens, and SG entry-value prediction | MSE and MAE | Helped |
| Use hierarchical Gamma processes with flexible location/sociability factor counts rather than NEST's restricted factor configurations | Ours versus NEST · Alog and SG link prediction | AUC | Mixed |
| Use hierarchical Gamma processes with flexible location/sociability factor counts rather than NEST's restricted factor configurations | Ours versus NEST · MovieLens link prediction | AUC | Helped |
| Model sparse entry indices rather than assuming a dense tensor | Link prediction · Alog, MovieLens, and SG | AUC | Helped |
| Use flexible sparse factor modeling instead of dense modeling | Prediction · SG compared with Alog and MovieLens | AUC | Helped |
| Set the number of location and sociability factors to different combinations while keeping total factors at 11 (ablation) | Ours · Alog and SG | MSE and MAE | Mixed |
| Choose the application-specific location/sociability factor balance (ablation) | Ours · Alog entry-value prediction | MSE and MAE | Helped |
| Choose the application-specific location/sociability factor balance (ablation) | Ours · SG entry-value prediction | MSE | Helped |
| Choose the application-specific location/sociability factor balance (ablation) | Ours · SG entry-value prediction | MAE | Helped |
| Increase the total number of location and sociability factors from the smaller tested settings to R=7, 9, or 11 (ablation) | Ours versus NEST · Alog and MovieLens entry-value prediction | MSE and MAE | Helped |
| Use our method rather than NEST at R=3 or R=5 (ablation) | Entry-value prediction · Alog, MovieLens, and SG | MSE and MAE | Hurt |
Closest papers by meaning
- Nonparametric Decomposition of Sparse Tensors (ICML 2021 · similarity 0.94)
- Gaussian process nonparametric tensor estimator and its minimax optimality (ICML 2016 · similarity 0.90)
- Efficient Nonparametric Tensor Decomposition for Binary and Count Data (AAAI 2024 · similarity 0.90)
- Self-Modulating Nonparametric Event-Tensor Factorization (ICML 2020 · similarity 0.89)
- High-Dimensional Tensor Regression With Oracle Properties (ICML 2025 · similarity 0.87)
- Performance Gaps in Multi-view Clustering under the Nested Matrix-Tensor Model (ICLR 2024 · similarity 0.87)
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.