Meta-Learning for Symbolic Hyperparameter Defaults
This addresses the need for efficient hyperparameter tuning in machine learning, though it is incremental as it builds on existing meta-learning and evolutionary methods.
The paper tackles the problem of hyperparameter optimization by proposing a zero-shot method to meta-learn symbolic default hyperparameter configurations based on dataset properties, enabling faster configuration than standard approaches. The method was evaluated on surrogate models and real data across 6 ML algorithms on over 100 datasets, demonstrating it finds viable symbolic defaults.
Hyperparameter optimization in machine learning (ML) deals with the problem of empirically learning an optimal algorithm configuration from data, usually formulated as a black-box optimization problem. In this work, we propose a zero-shot method to meta-learn symbolic default hyperparameter configurations that are expressed in terms of the properties of the dataset. This enables a much faster, but still data-dependent, configuration of the ML algorithm, compared to standard hyperparameter optimization approaches. In the past, symbolic and static default values have usually been obtained as hand-crafted heuristics. We propose an approach of learning such symbolic configurations as formulas of dataset properties from a large set of prior evaluations on multiple datasets by optimizing over a grammar of expressions using an evolutionary algorithm. We evaluate our method on surrogate empirical performance models as well as on real data across 6 ML algorithms on more than 100 datasets and demonstrate that our method indeed finds viable symbolic defaults.