.png){kind=icon}
.png){kind=icon}
Return
Support tensor ring kernel machine with dual-stage acceleration
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2026.113100.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose a new Support Tensor Ring Kernel Model (STRKM) by seamlessly coupling TR decomposition with kernel learning. • A new fast TR decomposition method is designed and a theoretical analysis of its approximation error bound is provided. • We construct a safe screening rule for STRKM to accelerate the model optimization. • To our knowledge, this is the first attempt to achieve holistic acceleration by jointly optimizing the tensor decomposition and model solution.
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W