Fast Construction of Correcting Ensembles for Legacy Artificial Intelligence Systems: Algorithms and a Case Study
This provides a computationally efficient method for enhancing generic AI systems, including neural networks, by correcting errors without major redesign, though it appears incremental as it builds on existing architectures.
The paper tackles the problem of improving legacy AI systems by constructing small network ensembles on top of existing architectures, enabling instantaneous removal of spurious and systematic errors with high probability on exponentially large datasets, as demonstrated in a case study on American Sign Language digit recognition.
This paper presents a technology for simple and computationally efficient improvements of a generic Artificial Intelligence (AI) system, including Multilayer and Deep Learning neural networks. The improvements are, in essence, small network ensembles constructed on top of the existing AI architectures. Theoretical foundations of the technology are based on Stochastic Separation Theorems and the ideas of the concentration of measure. We show that, subject to mild technical assumptions on statistical properties of internal signals in the original AI system, the technology enables instantaneous and computationally efficient removal of spurious and systematic errors with probability close to one on the datasets which are exponentially large in dimension. The method is illustrated with numerical examples and a case study of ten digits recognition from American Sign Language.