Bayesian Optimization with Shape Constraints
This work addresses the need for more efficient optimization in machine learning and decision-making by leveraging prior shape information, though it is incremental as it builds on existing Bayesian optimization frameworks.
The paper tackles the problem of incorporating shape constraints into Bayesian optimization, showing that it yields positive results in hyperparameter tuning and decision analysis.
In typical applications of Bayesian optimization, minimal assumptions are made about the objective function being optimized. This is true even when researchers have prior information about the shape of the function with respect to one or more argument. We make the case that shape constraints are often appropriate in at least two important application areas of Bayesian optimization: (1) hyperparameter tuning of machine learning algorithms and (2) decision analysis with utility functions. We describe a methodology for incorporating a variety of shape constraints within the usual Bayesian optimization framework and present positive results from simple applications which suggest that Bayesian optimization with shape constraints is a promising topic for further research.