-
#### Summary:
The number of log prob gradient evaluations per sample is one greater than the reported `n_leapfrog` for that sample.
This does not need to be so; the gradient for the starting point w…
-
Potential alternative to MH, supposed to be better.
A variant is the "NUTS" algorithm, which is implemented in the package "stan"
Should look into at some point.
-
- [ ] Can utilize the _or_ functionality e.g `string(a|b)string`.
- [ ] Can utilize the `*` ,`+` and `?` to indicate zero or more occurrences and atleast one occurrence and zero one occurrence respect…
-
## 一言でいうと
Annealed importance sampling (AIS)とHamiltonian Monte Carlo (HMC)を合体させた手法（Differentiable AIS, DAIS）を提案することで、AISによる周辺尤度の推定量が微分不可能となる問題を回避した。
### 論文リンク
[https://proceedings.neurips.cc/pape…
-
The `{mvgam}` package can now fit flexible latent variable models (including JSDMs) using the new [`jsdgam()` function](https://nicholasjclark.github.io/mvgam/reference/jsdgam.html). Would it be possi…
-
This came from : https://github.com/stan-dev/stan/wiki/Stan-Best-Practices/
It should be included in the user's guide in some form merged with what's there now about Stan progams as software.
On…
-
I wonder what you think about adaptive HMC, in particular [Wang, Mohamed, and Freitas (2013)](http://jmlr.org/proceedings/papers/v28/wang13e.pdf). Note that I am not asking you to implement it, just c…
tpapp updated
6 years ago
-
Hey, I worked briefly on this project a few years back looking at some ABC methods but essentially ran out of time and the code never made it to a production standard and performed fairly poorly. I've…
-
This is more of a long-term feature request. Sorry for rambling. TL;DR: dagger needs HMC.
The LMH algorithm works by sampling individual elementary stochastic events from their respective priors, t…
-
I'm opening this issue to begin a conversation about upgrading BayesianDataFusion.jl to be compatible with the most recent version of Julia, 1.x. We've begun upgrading the package on a fork and have s…