Baseline Results for Selected Nonlinear System Identification Benchmarks
This work provides a reference for researchers to benchmark new methods against established ones, though it is incremental as it compiles existing techniques without introducing new approaches.
The paper tackles the challenge of comparing nonlinear system identification methods by presenting baseline results from ten established techniques on five popular benchmarks, aiming to facilitate objective evaluation.
Nonlinear system identification remains an important open challenge across research and academia. Large numbers of novel approaches are seen published each year, each presenting improvements or extensions to existing methods. It is natural, therefore, to consider how one might choose between these competing models. Benchmark datasets provide one clear way to approach this question. However, to make meaningful inference based on benchmark performance it is important to understand how well a new method performs comparatively to results available with well-established methods. This paper presents a set of ten baseline techniques and their relative performances on five popular benchmarks. The aim of this contribution is to stimulate thought and discussion regarding objective comparison of identification methodologies.