Extraction of piecewise-linear analog circuit models from trained neural networks using hidden neuron clustering

This paper presents a new technique for automatically creating analog circuit models. The method extracts - from trained neural networks-piecewise linear models expressing the linear dependencies between circuit performances and design parameters. The paper illustrates the technique for an OTA circuit for which models for gain and bandwidth were automatically generated. The extracted models have a simple form that accurately fits the sampled points and the behavior of the trained neural networks. These models are useful for fast simulation of systems with non-linear behavior and performances.