Evolving machine learning workflows through interactive AutoML

arXiv:2402.18505v1 Announce Type: new
Abstract: Automatic workflow composition (AWC) is a relevant problem in automated machine learning (AutoML) that allows finding suitable sequences of preprocessing and prediction models together with their optimal hyperparameters. This problem can be solved using evolutionary algorithms and, in particular, grammar-guided genetic programming (G3P). Current G3P approaches to AWC define a fixed grammar that formally specifies how workflow elements can be combined and which algorithms can be included. In this paper we present \ourmethod, an interactive …

abstract algorithms arxiv automated automated machine learning automl cs.lg current evolutionary algorithms genetic programming grammar interactive machine machine learning machine learning workflows prediction prediction models programming through together type workflow workflows