A warped kernel improving robustness in Bayesian optimization via random embeddings
This work addresses robustness issues in Bayesian optimization for high-dimensional problems, but it is incremental as it builds on an existing approach.
The paper tackles the problem of high extrinsic dimensionality in Bayesian optimization by introducing a warped kernel within the Random Embedding Bayesian Optimization approach, resulting in improved robustness and alleviated constraints on bound selection, as demonstrated in a test case with 25 variables and intrinsic dimension 6.
This works extends the Random Embedding Bayesian Optimization approach by integrating a warping of the high dimensional subspace within the covariance kernel. The proposed warping, that relies on elementary geometric considerations, allows mitigating the drawbacks of the high extrinsic dimensionality while avoiding the algorithm to evaluate points giving redundant information. It also alleviates constraints on bound selection for the embedded domain, thus improving the robustness, as illustrated with a test case with 25 variables and intrinsic dimension 6.