Directional Linear Separability of Neural Representations: Geometry and Transformations
The directional linear separability measure (D-LSM) is introduced, which quantifies unavoidable competing-sample intrusion over affine halfspaces retaining every target sample, characterize its supporting geometry, and prove invariance under injective affine embeddings.
Yi Wei, Xuan Qi, Suorong Yang et al.
· 0 citations