Landmark Selection
Cross-source consensus on Landmark Selection from 1 sources and 4 claims.
1 sources · 4 claims
Uses
Benefits
Dosage & preparation
Comparisons
Highlighted claims
- The best CIFAR10 quality-runtime trade-off was reported for 128 to 512 landmarks per class. — DriftXpress: Faster Drifting Models via Projected RKHS Fields
- Density-aware and random landmark selection methods worked well on CIFAR10, while k-center underperformed. — DriftXpress: Faster Drifting Models via Projected RKHS Fields
- The paper identifies random per-class Nyström landmarks as a strong default. — DriftXpress: Faster Drifting Models via Projected RKHS Fields
- Random per-class landmark selection was much faster and competitive with global k-means. — DriftXpress: Faster Drifting Models via Projected RKHS Fields