Parameter Estimation on Synthetic p=q=200, M=20, High noise
0.08MSE_WSLM-Manifold
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SLM-ManifoldDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.08 | 0.06 | 0.37 | 0.84 | 0.26 | |
| BCD-SLMDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.08 | 0.06 | 0.37 | 0.84 | 0.26 | |
| SLM-OracleDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.08 | 0.06 | 0.36 | 0.81 | 0.22 | |
| EMDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.09 | 0.06 | 9.89 | 5.47 | 5 | |
| ECMDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.09 | 0.06 | 6.19 | 2.6 | 3.52 | |
| SLM-InteriorDimension p=200, Dimension q=200, Latent variables M=20, Rank r=5, Sample size N=2000, Noise level=High, Evaluation protocol=5-fold outer CV2026.05 | 0.22 | 0.17 | 1.64 | 1.17 | 0.45 |