Abstract:The trace quotient problem or the trace ratio problem (TRP) is to find an orthogonal matrix \(V\in \mathbb{R }^{m\times \ell } ~(m\ge \ell )\) that maximizes the quotient \({{\mathrm{tr}}}(V^\top BV)/{{\mathrm{tr}}}(V^\top WV)\) for a given symmetric matrix \(B\in \mathbb R ^{m\times m}\) and a symmetric positive definite matrix \(W\in \mathbb R ^{m\times m}.\) It has a crucial role in linear discriminant analysis and has many other applications in computer vision and machine learning as well. In this short note, we first establish the classical first and second order optimality conditions for TRP. As a straightforward application of these optimality conditions, we contribute a simple proof for the property that TRP does not admit local non-global maximizer, which is first proved by Shen et al. (A geometric revisit to the trace quotient problem, proceedings of the 19th International Symposium of Mathematical Theory of Networks and Systems, 2010) based on Grassmann manifold. Without involving much knowledge of the underlying differential geometry, our proof primarily uses basic properties in linear algebra, which also leads to an effective starting pointing strategy for any monotonically convergent iteration to find the global maximizer of TRP.

On Generalizing Trace Minimization Principles, II

On Generalizing Trace Minimization Principles, II

A Note on the Trace Quotient Problem.

Refined von Neumann-type trace inequality and its application

Double-variable trace maximization for extreme generalized singular quartets of a matrix pair: A geometric method

Extensions of some matrix inequalities related to trace and partial traces

Some convexity and monotonicity results of trace functionals

The Minimal Rank of Matrix Expressions with Respect to Hermitian Matrix-Revised

Closing the duality gap of the generalized trace ratio problem

Trace Formula For Two Variables

New results on the local-nonglobal minimizers of the generalized trust-region subproblem

Partial trace inequalities for partial transpose of positive semidefinite block matrices

The Mixed Second Absorption Laws of the Sum of Two Matrices on Generalized Inverse

Trace reconstruction of matrices and hypermatrices

Lower Bounds of the Smallest Singular Value of Matrices

Equalities and inequalities for Hermitian solutions and Hermitian definite solutions of the two matrix equations $AX = B$ and $AXA^* = B$

Perturbation Analysis for the Trace Quotient Problem

A Generalization of Lieb concavity theorem

Orthogonal Trace-Sum Maximization: Tightness of the Semidefinite Relaxation and Guarantee of Locally Optimal Solutions

Some new minimax theorems for generalized convexity

Generalized High-Dimensional Trace Regression via Nuclear Norm Regularization