Practical Bayesian Optimization for Variable Cost Objectives
This work addresses optimization efficiency for researchers and practitioners dealing with costly evaluations, though it appears incremental as it builds on existing Bayesian Optimization frameworks.
The paper tackles the problem of optimizing black-box functions with variable evaluation costs by proposing a Bayesian Optimization approach that balances cost and fidelity, achieving lower overhead than previous methods on synthetic and real-world benchmarks.
We propose a novel Bayesian Optimization approach for black-box functions with an environmental variable whose value determines the tradeoff between evaluation cost and the fidelity of the evaluations. Further, we use a novel approach to sampling support points, allowing faster construction of the acquisition function. This allows us to achieve optimization with lower overheads than previous approaches and is implemented for a more general class of problem. We show this approach to be effective on synthetic and real world benchmark problems.