Parameter estimation on Synthetic p=q=500, M=10, Low noise
0.01MSE (W)SLM-Manifold
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SLM-ManifoldDimension p=500, Dimension q=500, Latent variables M=10, Rank r=5, Sample size N=2000, Noise level=Low, Evaluation protocol=5-fold outer CV2026.05 | 0.01 | 0.01 | 0.22 | 0.3 | 0.3 | |
| BCD-SLMDimension p=500, Dimension q=500, Latent variables M=10, Rank r=5, Sample size N=2000, Noise level=Low, Evaluation protocol=5-fold outer CV2026.05 | 0.01 | 0.01 | 0.22 | 0.3 | 0.3 | |
| EMDimension p=500, Dimension q=500, Latent variables M=10, Rank r=5, Sample size N=2000, Noise level=Low, Evaluation protocol=5-fold outer CV2026.05 | 0.01 | 0.01 | 0.17 | 0.14 | 0 | |
| ECMDimension p=500, Dimension q=500, Latent variables M=10, Rank r=5, Sample size N=2000, Noise level=Low, Evaluation protocol=5-fold outer CV2026.05 | 0.01 | 0.02 | 0.19 | 0.15 | 0 | |
| SLM-InteriorDimension p=500, Dimension q=500, Latent variables M=10, Rank r=5, Sample size N=2000, Noise level=Low, Evaluation protocol=5-fold outer CV2026.05 | 0.06 | 0.1 | 5.06 | 4.7 | 4 |