Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Diversified Bayesian Nonnegative Matrix Factorization
Qiao Maoying, Yu Jun, Liu Tongliang, Wang Xinchao, Tao Dacheng
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 DPP diversity prior (DivBNMF-MAP) instead of plain NMF | NMF · Synthetic K=2, Euclidean | reconstruction error: 25.68 -> 24.10 | Improved |
| add DPP diversity prior (DivBNMF-MAP) instead of plain NMF | NMF · Synthetic K=8, Frobenius | reconstruction error: 23.11 -> 21.09 | Improved |
| DivBNMF-MAP versus LVNMF | large-volume NMF · Synthetic K=6, Euclidean | reconstruction error: 20.07 -> 17.74 | Improved |
| DivBNMF-MAP versus LANMF | large-angle NMF · Synthetic K=4, Euclidean | reconstruction error: 21.40 -> 19.55 | Improved |
| MAP estimate instead of posterior mean in DivBNMF | DivBNMF · Synthetic K=8, Euclidean | reconstruction error: 15.23 -> 14.67 | Improved |
| MAP estimate instead of posterior mean in DivBNMF | DivBNMF · Synthetic K=10, Frobenius | reconstruction error: 0.23 -> 0.22 | No clear effect |
| increase number of bases K | NMF · Synthetic, Euclidean | reconstruction error: 25.68 -> 16.57 | Improved |
| DivBNMF at K=10 versus diversity baselines | DivBNMF-MAP · Synthetic K=10, Euclidean | reconstruction error: 0.08 -> 0.09 | Mixed |
| use diversity-encouraging priors/regularizers instead of NMF | NMF · MULAN scene, multi-label learning | average precision | Improved |
| DivBNMF-MAP versus baselines per class | DivBNMF-MAP · MULAN scene classes 1-4 | average precision | Improved |
| DivBNMF-MAP versus baselines per class | DivBNMF-MAP · MULAN scene classes 5-6 | average precision | No clear effect |
| diversity prior on reconstruction error | DivBNMF-MAP · MULAN scene train/test | reconstruction error | No clear effect |
Closest papers by meaning
- Distributionally Robust and Multi-Objective Nonnegative Matrix Factorization (TPAMI 2021 · similarity 0.89)
- Generalized Separable Nonnegative Matrix Factorization (TPAMI 2019 · similarity 0.89)
- A Particle-Based Variational Approach to Bayesian Non-negative Matrix Factorization (JMLR 2019 · similarity 0.88)
- An Exterior Method for Nonnegative Matrix Factorization (ICML 2026 · similarity 0.87)
- Prior Specification for Bayesian Matrix Factorization via Prior Predictive Matching (JMLR 2023 · similarity 0.86)
- Statistically Optimal $K$-means Clustering via Nonnegative Low-rank Semidefinite Programming (ICLR 2024 · similarity 0.84)
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.
