Fix defaults and suggestions

[aleximmer]

In contrast to the redux paper, we give an example with post-hoc Laplace and marginal likelihood prior precision tuning in the docs (https://aleximmer.github.io/Laplace/#full-example-post-hoc-laplace-on-a-large-image-classifier). This is confusing and we have not thoroughly tested this approach but only recommended full + online marglik or last-layer + validation-based prior tuning.

[wiseodd]

For sequential decision making (BO & contextual bandits), there doesn't seem to be any difference between post-hoc and online marglik, in my experience. See

• https://arxiv.org/abs/2310.00137
• https://arxiv.org/abs/2304.08309

I also observed this in my latest BO-with-LLM paper, although the results are not published.

I think it's a matter of putting this as a caveat in the README.

[aleximmer]

Makes sense. I think it's just a matter of generally making sure that people don't expect everything to work automatically and be optimal but rather try different options depending on the problem.