Transfer Learning for Bayesian Optimization on Heterogeneous Search Spaces
This work addresses a bottleneck in Bayesian optimization for researchers and practitioners by enabling transfer learning across different domains, though it is incremental as it builds on existing transfer learning approaches.
The paper tackled the problem of transferring knowledge across heterogeneous search spaces in Bayesian optimization, introducing MPHD, a model pre-training method that uses neural networks to map domain-specific contexts to hierarchical Gaussian processes, and demonstrated its superior performance on challenging black-box function optimization tasks.
Bayesian optimization (BO) is a popular black-box function optimization method, which makes sequential decisions based on a Bayesian model, typically a Gaussian process (GP), of the function. To ensure the quality of the model, transfer learning approaches have been developed to automatically design GP priors by learning from observations on "training" functions. These training functions are typically required to have the same domain as the "test" function (black-box function to be optimized). In this paper, we introduce MPHD, a model pre-training method on heterogeneous domains, which uses a neural net mapping from domain-specific contexts to specifications of hierarchical GPs. MPHD can be seamlessly integrated with BO to transfer knowledge across heterogeneous search spaces. Our theoretical and empirical results demonstrate the validity of MPHD and its superior performance on challenging black-box function optimization tasks.