botorch
botorch copied to clipboard
pytorch
Bayesian optimization in PyTorch
A series of improvements directed towards improving the performance of PES & JES, as well as their MultiObj counterparts. ## Motivation As pointed out by @SebastianAment in [this paper](https://arxiv.org/pdf/2310.20708), the...
## Motivation Fixes #2740 @sdaulton: This is a draft for fixing the dimensionality issue. While the PR technically solves the problem, think it only addresses the symptom and not the...
### What happened? Not sure if this is intentional, but it seems like a bug to me. When fixing all features in `optimize_acqf`, the returned acquisition value changes from a...
### Motivation Would it be possible to make `StratifiedStandardize` more general so that it can work with multi-output models as `Standardize`? This would be needed for `MultiTaskGP` which requires stratified...
### Motivation TabPFN2 shows impressive results, also for Regression (https://github.com/PriorLabs/TabPFN). The model is fully differentiable and can also quantify its uncertainty, so it should be also suitable for BO (there...
## Motivation I wanted to experiment with the vizier output transforms in botorch - in particular the infeasible transform. Potentially misplaced effort as it appears that within the meta ecosystem...
# 🐛 Bug In the documentation for BoTorch 0.11.3, I-BNN is shown to outperform standard GPs with Matern and RBF kernels: https://botorch.org/v/0.11.3/tutorials/ibnn_bo In the 0.12.0 documentation, the Matern kernel's performance...
# 🐛 Bug When using `nonlinear_inequality_constraints` in `optimize_acqf`, you need to set `batch_initial_conditions`, and these ICs need to respect the constraints. This seems unnecessary - SLSQP is capable of starting...
## Motivation Since version 0.12.0, dim_scaled_lognormal_prior[Hvarfner2024vanilla] has become the default. However, as ScaleKernel is not applied in `get_covar_module_with_dim_scaled_prior()`, performance may deteriorate in some cases. (The selection of the prior distribution...
# 🚀 Feature Request The `Log` outcome transform currently raises a `NotImplementedError` if it is used with known observation noise. This limits its usability. The transform should be updated to...
