General-purpose hierarchical optimisation of machine learning pipelines with grammatical evolution

This paper introduces Hierarchical Machine Learning Optimisation (HML-Opt), an AutoML framework that is based on probabilistic grammatical evolution. HML-Opt has been designed to provide a flexible where researcher can define the space of possible pipelines solve specific machine learning problem, which range from high-level decisions about representation and features low-level hyper-parameter values. The evaluation presented via two different case studies, both demonstrate it competitive with existing tools variety benchmarks. Furthermore, be applied novel problems, such as knowledge extraction natural language text, whereas other techniques are insufficiently capture complexity these scenarios. source code for available online research community.