Informative Laplacian Projection
نویسندگان
چکیده
A new approach of constructing the similarity matrix for eigendecomposition on graph Laplacians is proposed. We first connect the Locality Preserving Projection method to probability density derivatives, which are then replaced by informative score vectors. This change yields a normalization factor and increases the contribution of the data pairs in low-density regions. The proposed method can be applied to both unsupervised and supervised learning. Empirical study on facial images is provided. The experiment results demonstrate that our method is advantageous for discovering statistical patterns in sparse data areas.
منابع مشابه
Abelian Projection without Ambiguities
Laplacian Abelian Projection is discussed. This term refers to the use of a new (“Laplacian”) gauge fixing prescription for implementing the Abelian Projection of QCD. The gauge condition is based on the lowest-lying eigenvector of the covariant Laplacian operator in the adjoint representation. This Laplacian gauge fixing procedure is free of the ambiguities which plague lattice simulations whi...
متن کاملLaplacian Abelian Projection
A new partial gauge fixing condition for the abelian projection is introduced. It is based on the lowest-lying eigenvector of a covariant Laplacian operator. This gauge is smooth and free of lattice Gribov copies. These properties are important for an unambiguous computation of the abelian projected gauge field configuration.
متن کاملConformal Powers of the Laplacian via Stereographic Projection
A new derivation is given of Branson’s factorization formula for the conformally invariant operator on the sphere whose principal part is the k-th power of the scalar Laplacian. The derivation deduces Branson’s formula from knowledge of the corresponding conformally invariant operator on Euclidean space (the k-th power of the Euclidean Laplacian) via conjugation by the stereographic projection ...
متن کاملGradient estimates for eigenfunctions on compact Riemannian manifolds with boundary
The purpose of this paper is to prove the L∞ gradient estimates and L∞ gradient estimates for the unit spectral projection operators of the Dirichlet Laplacian and Neumann (or more general, Ψ1-Robin) Laplacian on compact Riemannian manifolds (M, g) of dimension n ≥ 2 with C2 boundary . And we also get an upper bounds for normal derivatives of the unit spectral projection operators of the Dirich...
متن کاملLaplacian MinMax Discriminant Projection and its Applications
A new algorithm, Laplacian MinMax Discriminant Projection (LMMDP), is proposed in this paper for supervised dimensionality reduction. LMMDP aims at learning a discriminant linear transformation. Specifically, we define the within-class scatter and the between-class scatter using similarities which are based on pairwise distances in sample space. After the transformation, the considered pairwise...
متن کامل