A Warped Kernel Improving Robustness in Bayesian Optimization Via Random Embeddings
نویسندگان
چکیده
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.
منابع مشابه
Bayesian Optimization in High Dimensions via Random Embeddings
Bayesian optimization techniques have been successfully applied to robotics, planning, sensor placement, recommendation, advertising, intelligent user interfaces and automatic algorithm configuration. Despite these successes, the approach is restricted to problems of moderate dimension, and several workshops on Bayesian optimization have identified its scaling to high dimensions as one of the h...
متن کاملLearning Random Fourier Features by Hybrid Constrained Optimization
The kernel embedding algorithm is an important component for adapting kernel methods to large datasets. Since the algorithm consumes a major computation cost in the testing phase, we propose a novel teacher-learner framework of learning computation-efficient kernel embeddings from specific data. In the framework, the highprecision embeddings (teacher) transfer the data information to the comput...
متن کاملBayesian Optimization in a Billion Dimensions Bayesian Optimization in a Billion Dimensions via Random Embeddings
Bayesian optimization techniques have been successfully applied to robotics, planning, sensor placement, recommendation, advertising, intelligent user interfaces and automatic algorithm configuration. Despite these successes, the approach is restricted to problems of moderate dimension, and several workshops on Bayesian optimization have identified its scaling to high-dimensions as one of the h...
متن کاملBayesian Optimization in a Billion Dimensions via Random Embeddings
Bayesian optimization techniques have been successfully applied to robotics, planning, sensor placement, recommendation, advertising, intelligent user interfaces and automatic algorithm configuration. Despite these successes, the approach is restricted to problems of moderate dimension, and several workshops on Bayesian optimization have identified its scaling to high-dimensions as one of the h...
متن کاملA Bayesian Approach to Estimate Parameters of a Random Coefficient Transition Binary Logistic Model with Non-monotone Missing Pattern and some Sensitivity Analyses
A transition binary logistic model with random coefficients is proposed to model the unemployment statues of household members in two seasons of spring and summer. Data correspond to the labor force survey performed by Statistical Center of Iran in 2006. This model is introduced to take into account two kinds of correlation in the data one due to the longitudinal nature o...
متن کامل