TransBO: Hyperparameter Optimization via Two-Phase Transfer Learning
This addresses the problem of efficient hyperparameter tuning for machine learning practitioners, offering an incremental improvement by integrating transfer learning into existing HPO methods.
The paper tackles hyperparameter optimization by proposing TransBO, a two-phase transfer learning framework that leverages past tasks to accelerate current tuning, demonstrating superiority over state-of-the-art methods in experiments including neural architecture search.
With the extensive applications of machine learning models, automatic hyperparameter optimization (HPO) has become increasingly important. Motivated by the tuning behaviors of human experts, it is intuitive to leverage auxiliary knowledge from past HPO tasks to accelerate the current HPO task. In this paper, we propose TransBO, a novel two-phase transfer learning framework for HPO, which can deal with the complementary nature among source tasks and dynamics during knowledge aggregation issues simultaneously. This framework extracts and aggregates source and target knowledge jointly and adaptively, where the weights can be learned in a principled manner. The extensive experiments, including static and dynamic transfer learning settings and neural architecture search, demonstrate the superiority of TransBO over the state-of-the-arts.