Efficiency of Shallow Cascades for Improving Deep Learning AI Systems
Ivan Tyukin, Alexander N. Gorban, Danil V. Prokhorov, Stephen A. Green · 2018
This paper presents a technology for simple and non-iterative improvements of Multilayer and Deep Learning neural networks and Artificial Intelligence (AI) systems. The improvements are, in essence, shallow networks constructed on top of the existing Deep Learning architecture. Theoretical foundation of the technology is based on Stochastic Separation Theorems and the ideas of measure concentration. We show that, subject to mild technical assumptions on statistical properties of internal signals in Deep Learning AI, with probability close to one the technology enables instantaneous “learning away” of spurious and systematic errors. The method is illustrated with numerical examples.