Operator theory, kernels, and Feedforward Neural Networks
This work addresses theoretical analysis for neural network algorithms, but appears incremental as it applies existing kernel methods to specific network contexts.
The paper tackles the analysis of iteration algorithms for multi-layer feedforward neural networks using specific families of positive definite kernels, focusing on kernels that adapt to data with intrinsic self-similarities at scaling iterations.
In this paper we show how specific families of positive definite kernels serve as powerful tools in analyses of iteration algorithms for multiple layer feedforward Neural Network models. Our focus is on particular kernels that adapt well to learning algorithms for data-sets/features which display intrinsic self-similarities at feedforward iterations of scaling.