Label Distribution Learning on SBU (3 random splits)
0.0414KL DivergenceERDF
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| ERDFnumber of trees=100, max depth=10, max cascade forest layers=10, early stopping tolerance=1, feature enhancement k=52026.02 | 0.0414 | 0.0865 | 0.2746 | 0.5772 | 0.9587 | 89.68 | |
| StructRFnumber of trees=100, max depth=102026.02 | 0.0573 | 0.1104 | 0.3444 | 0.7276 | 0.9432 | 86.91 | |
| LDL-SCLhyperparameters=λ1, λ2, λ3 tuned from {10^-1, 10^-2, 10^-3}, number of clusters=5 or 102026.02 | 0.0634 | 0.1199 | 0.3695 | 0.796 | 0.9376 | 85.75 | |
| AA-KNNk=5, distance metric=minkowski2026.02 | 0.0802 | 0.1273 | 0.4012 | 0.8309 | 0.9219 | 84.87 | |
| AA-BPoptimizer=SA-BFGS, max iterations=5002026.02 | 0.1065 | 0.1378 | 0.5113 | 1.0997 | 0.9039 | 81.06 |