|
Generalized Spectral Bounds for Sparse LDA
Baback Moghadam, Yair Weiss and Shai Avidan
In: ICML 2006, Pittsburgh(2006).
AbstractWe present a discrete spectral framework for
the sparse or cardinality-constrained solution
of a generalized Rayleigh quotient. This NPhard
combinatorial optimization problem is
central to supervised learning tasks such as
sparse LDA, feature selection and relevance
ranking for classication. We derive a new
generalized form of the Inclusion Principle
for variational eigenvalue bounds, leading to
exact and optimal sparse linear discriminants
using branch-and-bound search. An ecient
greedy (approximate) technique is also presented.
The generalization performance of
our sparse LDA algorithms is demonstrated
with real-world UCI ML benchmarks and
compared to a leading SVM-based gene
selection algorithm for cancer classification. [Edit]
| 