Put CASH on Bandits: A Max K-Armed Problem for Automated Machine Learning

The Combined Algorithm Selection and Hyperparameter optimization (CASH) is a challenging resource allocation problem in the field of AutoML. We propose MaxUCB, a max $k$-armed bandit method to trade off exploring different model classes and conducting hyperparameter optimization. MaxUCB is specifically designed for the light-tailed and bounded reward distributions arising in this setting and, thus, provides an efficient alternative compared to classic max $k$-armed bandit methods assuming heavy-tailed reward distributions. We theoretically and empirically evaluate our method on four standard AutoML benchmarks, demonstrating superior performance over prior approaches.

@article{balef2025_2505.05226,
  title={ Put CASH on Bandits: A Max K-Armed Problem for Automated Machine Learning },
  author={ Amir Rezaei Balef and Claire Vernade and Katharina Eggensperger },
  journal={arXiv preprint arXiv:2505.05226},
  year={ 2025 }
}