Research map / Learning theory and algorithms / Matrix and tensor methods
Learning from Binary Multiway Data: Probabilistic Tensor Decomposition and its Statistical Optimality
Miaoyan Wang, Lexin Li
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 |
|---|---|---|---|
| use binary (logistic) tensor decomposition instead of continuous-valued decomposition | BTF_Alternating · Enron tensor completion | AUC: 79.6% -> 94.3% | Improved |
| use binary instead of continuous-valued tensor decomposition | BTF_Alternating · Kinship, Nations, HCP completion | AUC: 0.9436 -> 0.9708 | Improved |
| use binary instead of continuous-valued tensor decomposition | BTF_Alternating · Enron tensor completion | RMSE: 6.3 x 10^-5 -> 6.4 x 10^-3 | Hurt |
| increase tensor dimension d | Constrained MLE · Simulated CP tensor | estimation error | Improved |
| increase true rank R | Constrained MLE · Simulated CP tensor | estimation error | Hurt |
| increase noise level sigma in high SNR regime | Constrained MLE · Simulated CP tensor, d=50 | estimation error | Improved |
| increase dimension d from 20 to 40 for BIC rank selection | BIC · Rank selection | selected rank: 8.7 -> 10 | Improved |
| use probabilistic Bernoulli model rather than BooleanTF | BTF methods · Boolean tensor model data | RMSE | Improved |
| use BTF_Alternating or BTF_Gradient instead of BTF_Bayesian | BTF_Alternating · Boolean tensor model data | misclassification error | Improved |
| fit logistic link to probit-generated data (misspecification) | BTF_Alternating · Stochastic multiway block model | relative loss: 0.22 to 0.48 relative loss across sub-models | No clear effect |
| use rank-R CP exact low-rank instead of unfolding into matrix | MLE upper bound · Theory | convergence rate: d^-(K-1)/4 -> d^-(K-1)/2 | Improved |
Closest papers by meaning
- Tensor denoising and completion based on ordinal observations (ICML 2020 · similarity 0.90)
- Performance Gaps in Multi-view Clustering under the Nested Matrix-Tensor Model (ICLR 2024 · similarity 0.88)
- Multiway clustering via tensor block models (NeurIPS 2019 · similarity 0.88)
- On Tensor Train Rank Minimization : Statistical Efficiency and Scalable Algorithm (NeurIPS 2017 · similarity 0.88)
- Scaling and Scalability: Provable Nonconvex Low-Rank Tensor Estimation from Incomplete Measurements (JMLR 2022 · similarity 0.88)
- Efficient Nonparametric Tensor Decomposition for Binary and Count Data (AAAI 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.
