Consolidated learning -- a domain-specific model-free optimization strategy with examples for XGBoost and MIMIC-IV

TL;DR

This paper introduces consolidated learning, a domain-specific, model-free hyperparameter optimization strategy that uses a static portfolio of configurations to efficiently tune multiple similar models, demonstrated on XGBoost and MIMIC-IV data.

Contribution

It proposes a new formulation of hyperparameter tuning focused on multiple models and introduces a static portfolio approach for efficient optimization in similar data domains.

Findings

Static hyperparameter portfolios perform well for multiple models.

Transfer of hyperparameters is more efficient across similar tasks.

Effective in practical settings like XGBoost and MIMIC-IV.

Abstract

For many machine learning models, a choice of hyperparameters is a crucial step towards achieving high performance. Prevalent meta-learning approaches focus on obtaining good hyperparameters configurations with a limited computational budget for a completely new task based on the results obtained from the prior tasks. This paper proposes a new formulation of the tuning problem, called consolidated learning, more suited to practical challenges faced by model developers, in which a large number of predictive models are created on similar data sets. In such settings, we are interested in the total optimization time rather than tuning for a single task. We show that a carefully selected static portfolio of hyperparameters yields good results for anytime optimization, maintaining ease of use and implementation. Moreover, we point out how to construct such a portfolio for specific domains. The improvement in the optimization is possible due to more efficient transfer of hyperparameter configurations between similar tasks. We demonstrate the effectiveness of this approach through an empirical study for XGBoost algorithm and the collection of predictive tasks extracted from the MIMIC-IV medical database; however, consolidated learning is applicable in many others fields.