Flexible model composition in machine learning and its implementation in MLJ
This work provides a more flexible and robust method for model composition, which is a common challenge for machine learning practitioners building complex systems.
This paper introduces 'learning networks', a graph-based protocol for composing diverse machine learning models into meta-models. This approach addresses limitations found in current dominant ML platforms and is implemented with a concise syntax in the MLJ framework.
A graph-based protocol called `learning networks' which combine assorted machine learning models into meta-models is described. Learning networks are shown to overcome several limitations of model composition as implemented in the dominant machine learning platforms. After illustrating the protocol in simple examples, a concise syntax for specifying a learning network, implemented in the MLJ framework, is presented. Using the syntax, it is shown that learning networks are are sufficiently flexible to include Wolpert's model stacking, with out-of-sample predictions for the base learners.