A composite polynomial filter approximates PSD cone projection with matrix multiplications alone, achieving roughly 10x speedup over GPU eigenvalue solvers at about 1e-3 relative error.
Spectral operators of matrices.Mathematical Programming, 168:509–531, 2018
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
math.OC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Factorization-free Orthogonal Projection onto the Positive Semidefinite Cone with Composite Polynomial Filtering
A composite polynomial filter approximates PSD cone projection with matrix multiplications alone, achieving roughly 10x speedup over GPU eigenvalue solvers at about 1e-3 relative error.