Shuhei Watanabe

Hi @not522 , I checked those papers before: * Scott, D.W. (1992) Multivariate Density Estimation: Theory, Practice, and Visualization. * Berwin, A.T. (1993) Bandwidth Selection in Kernel Density Estimation: A...

@not522 I checked [my implementation](https://github.com/nabenabe0928/tpe/tree/single-opt) and the Optuna implementation on some **high-dimensional** (50d) [benchmark functions](https://github.com/nabenabe0928/tpe/blob/single-opt/tpe/utils/benchmarks.py) with `n_trials = 200`. I describe findings here: 1. Mine and Optuna implementation yield more...

I did additional exhaustive experiments (4096 parameter settings on 51 targets over 10 random seeds) using [this repository](https://github.com/nabenabe0928/tpe/tree/single-opt), so if you need preliminary results, feel free to ask me:) I...

```python Stack trace: 4 ~/.local/share/pypoetry/venv/lib/python3.9/site-packages/poetry/puzzle/solver.py:155 in _solve 153│ 154│ try: → 155│ result = resolve_version(self._package, self._provider) 156│ 157│ packages = result.packages 3 ~/.local/share/pypoetry/venv/lib/python3.9/site-packages/poetry/mixology/__init__.py:18 in resolve_version 16│ solver = VersionSolver(root, provider)...

Memo https://github.com/optuna/optuna/compare/master...nabenabe0928:refactor-optim-mixed-clean-with-class?expand=1

Additional dependencies for `optuna-integration` can now installed via pip and I modified the error message so that users can easily notice. With that, let me close this issue.

I will close this PR because this PR relates to #3450 .

Just in case, I am already working on a faster implementation of WFG. https://github.com/nabenabe0928/optuna/compare/enhance/speed-up-non-dominated-sort...nabenabe0928:optuna:enhance/speed-up-wfg

As mentioned [here](https://github.com/optuna/optuna/pull/5504#issuecomment-2196548251), hypervolume computation for 3D or more requires a special data structure. So if anyone would like to contribute to this issue, please note that it may take...

Sorry for the late response, I completely missed the thread. It is already kinda late, but I add some comments from my side in terms of the quantification itself: So...