I’m trying to speed up model convergence, and am considering finding the posterior mode using Stan’s LBFGS routines, and then initializing the sampler chains with it, or a perturbation. At first glance, I would expect this to be a relatively standard thing to do. However, I couldn’t find any inform…

[image] Todor: Are there in fact advantages in initializing the sampler close to the mode? What are the drawbacks and caveats one should be aware of? I would say it is not worthwhile. The drawback that a lot of people overlook is that for models that are so complicated that the default initia…

What Ben said, but for some models it works in the Craig’s-list-used-car sense and might give you a hint if you’re stuck on why your initial values are not working. So no reason not to try it if you can’t find alternatives

I think you are both misunderstanding my question. It’s more fundamental rather than about a specific model. The default values work. What I’m missing is, why would they work in any way better than the mode? Why isn’t the mode the default? The particular case that made me think about this: I have a…

When @bgoodri said [image] bgoodri: the region around the mode can have essentially zero posterior probability he was referring to something called concentration of measure . And you should heed his warning: if your model is “big” (I understood this as high dimensional, correct me if I’m wron…

In terms of advantages to initialising the sampler close to the mode, it seems to work well for clustering models. In the Stan User Guide, under the Clustering Models -> Multimodality section, it mentions: “the advice often given in fitting clustering models is to try many different initializations…

I’m a bit late to this but you might want to browse this thread where I proposed the same idea. Some good discussion: [image] Overdispersed initial values -- general questions Algorithms I know the general advice is to initialize Stan (or any MCMC algorithm) from widely d…

Using output from optimization algorithms to initialize sampler

Algorithms

monnahc April 25, 2019, 6:51pm 7

I’m a bit late to this but you might want to browse this thread where I proposed the same idea. Some good discussion:

Topic		Replies	Views
Overdispersed initial values -- general questions Algorithms ecology	20	4532	May 26, 2018
Is it fair/legitimate to use least square or optimization estimated parameter values to initialize the chain? Modeling techniques	6	828	April 8, 2021
Do priors influence initial values? Modeling	7	1693	September 24, 2018
Why doesn't sampling fail when optimizing does? Algorithms optimization , mcmc	3	925	February 13, 2020
Undesirable(?) behavior in initial values of hierarchical model General	9	1005	November 13, 2020

Using output from optimization algorithms to initialize sampler

Related topics