.png){kind=icon}
Return
Adaptive kernel learning using genetic programming in reproducing kernel banach spaces
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2026.116522.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel framework (Evo-RK) automates kernel construction in Reproducing Kernel Banach Spaces.
• Genetic Programming co-evolves nonlinear feature mappings and geometric parameters.
• The approach transcends the symmetric constraints of traditional Reproducing Kernel Hilbert Spaces.
• Evolved kernels achieve superior classification accuracy and significant model sparsity.
• “Deep kernel” architectures are autonomously discovered, enhancing interpretability.
Keywords:
Genetic programming
Reproducing kernel banach space
Semi-inner product
Support vector machine
Kernel method
Model sparsity
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
Cited Papers
No cited papers available