Inducing Points and Variational Inference in Gaussian Process
Bottleneck of computing Exact Gaussian Process Suppose we have n-dimensional Gaussian Process below which consist of n number of data.…
MachineLearning
kevin-tofu SingleTaskGP vs ModelListGP with SingleTaskGPs
About In BoTorch documentation, it is clearly mentioned that multiple-output and multiple task is different, and every GPs including SingleTaskGP…
Sparse Axis Aligned Subspace Bayesian Optimization
About Bayesian optimization using Gaussian processes is an extremely powerful tool. However, it is well-known that performance can degrade significantly…
Log Normal and Half-Cauchy Distribution
About High-Dimensional Bayesian Optimization with Sparse Axis-Aligned Subspaces, so called SAASBO (https://arxiv.org/abs/2103.00349) uses Log Normal distribution and Half-Cauchy Distribution on…
The feature of Stochastic Variational Inference
About https://arxiv.org/abs/1206.7051 The main goal is to perform Bayesian inference efficiently and scalable for large data sets and complex Bayesian…
Multi-Objective Optimization Problem in BoTorch
About There are some of functions for multi-objective optimization in BoTorch as toy problem. They inherit MultiObjectiveTestProblem class below which…
Expected Hypervolume Improvement in BoTorch
About Understanding EHVI is quite complicated. So by following the BoTorch code and some papers, I will quickly see the…
Optimize Gaussian Processes in BoTorch
About In BoTorch, there are 2 fitting functions are prepared, fit_gpytorch_mll / fit_fully_bayesian_model_nuts(https://botorch.org/api/fit.html). And there are some loss functions are…
The difference between Kernel PCA and GPLVM
About The Kernel PCA and the GPLVM, both are using Kernel function and the purpose of these are almost same,…
Focal Loss, Implementation in PyTorch
About https://arxiv.org/abs/1708.02002 The Focal Loss is a loss function designed to solve class imbalance problems, especially when some classes appear…