HEBO: Pushing The Limits of Sample-Efficient Hyperparameter Optimisation

Alexander Imani Cowen-Rivers, Wenlong Lyu, Rasul Tutunov, Zhi Wang, Antoine Grosnit, Ryan-Rhys Griffiths, Alexandre Max Maraval, Jianye HAO, Jun Wang, Jan Peters, Haitham Bou Ammar

Link To Paper: https://jair.org/index.php/jair/article/view/13643

Journal Of Paper: JAIR

Confirmed Open Access: Yes

Topics From Call For Papers:

Hyperparameter Optimization (HPO)
Bayesian Optimization for AutoML
Evolutionary Algorithms for AutoML
Multi-Objective Optimization for AutoML

Broader Impact Statement On Ethical And Societal Implications: Whilst we optimise models for downstream tasks, we do not consider controlling/ preventing the biases learnt from the machine learning models. We urge practitioners to always perform an in-depth analysis of the features used for machine learning models and, particularly when models being optimised input sensitive information, attend to sensitive information accordingly.