-
Approximations allow us to include model components that don't have autodiff-friendly implementations.
- A first test case will be approximating the local coupling term in a neural field model
- A…
-
I don't know anything about the implementation of `norm_num`, but in my ideal world, these lemmas would entirely be handled by `norm_num`.
_Originally posted by @Smaug123 in https://github.com/lean…
-
https://www.sciencedirect.com/science/article/pii/S0377042706004195 is a good reference.
-
-
Bootstrapped synthetic likelihood could prevent the repeated N times of fitting; resampling from one set of recovered parameters with some estimated variance could be used to approximate SBC_vanila's …
-
Is there any function computing Laplace approximation (similar to ```ed.Laplace```) supported at tensorflow probability at the moment?
-
the inverse of the curvature matrix is not a symmetric definite positive
-
We can currently do:
intvar = MixedIntervening(problem.n, problem.m + 1, default=Linear())
intvar.set_intervening(MMA(), var=1)
approx = Taylor1(intvar)
subproblem = Subproblem(a…
-
I'd consider these exact:
```
base(.01,sqrt 5)
.2 to base sqrt 5
```
Is it just because base conversions are approximate by default? These calculations can be optimized, IMHO.
-
`GetProjectilesInRectangle` exists, but not `GetProjectilesInCircle`, so if you want projectiles within a circle you need to use a square approximation.