A meta-learning recommender system for hyperparameter tuning: predicting when tuning improves SVM classifiers
This work addresses the computational cost of hyperparameter tuning for machine learning practitioners, offering an incremental improvement by optimizing when to tune.
The paper tackles the problem of predicting when hyperparameter tuning significantly improves SVM classifier performance, using a meta-learning recommender system to decide between default values and tuning. Results show accurate predictions across 156 datasets, reducing optimization time without compromising performance.
For many machine learning algorithms, predictive performance is critically affected by the hyperparameter values used to train them. However, tuning these hyperparameters can come at a high computational cost, especially on larger datasets, while the tuned settings do not always significantly outperform the default values. This paper proposes a recommender system based on meta-learning to identify exactly when it is better to use default values and when to tune hyperparameters for each new dataset. Besides, an in-depth analysis is performed to understand what they take into account for their decisions, providing useful insights. An extensive analysis of different categories of meta-features, meta-learners, and setups across 156 datasets is performed. Results show that it is possible to accurately predict when tuning will significantly improve the performance of the induced models. The proposed system reduces the time spent on optimization processes, without reducing the predictive performance of the induced models (when compared with the ones obtained using tuned hyperparameters). We also explain the decision-making process of the meta-learners in terms of linear separability-based hypotheses. Although this analysis is focused on the tuning of Support Vector Machines, it can also be applied to other algorithms, as shown in experiments performed with decision trees.