Lena

Stochastically constrained nodes can lead to wrong stationary distribution

1

This has to do with many of the test cases in [test_constrain.py](https://github.com/probcomp/Venturecxx/blob/release-0.4.2/test/conformance/test_constrain.py). In particular, consider the program: ``` (assume x (mem (lambda (i) (normal 0 1)))) (assume xs (mapv x...

AEInfer happens regardless of the inference program, except when it doesn't

3

As far as I can tell, AEinfer is called here: https://github.com/probcomp/Venturecxx/blob/master/backend/lite/trace.py#L543 In particular: - It happens once per transition of any primitive inference operator, including weird ones like `draw_scaffold` and...

global_log_likelihood does not include constrained choices with no random choices behind them

3

``` venture[script] > observe (normal 0 1) 10 venture[script] > global_log_likelihood [0] ``` This is because global_log_likelihood is implemented in terms of regen, and regen only ever cares about things...

Potential Venstan example? "double" mixture model

6

See: https://webfiles.uci.edu/mdlee/LeeWagenmakers2013_Free.pdf ("6.4 The two country quiz") https://groups.google.com/forum/#!topic/stan-users/p2zWntwTbG0 - Some context: all of the models in the Lee & Wagenmakers book were made into Stan examples that are linked from...

Slice sampling with a fixed interval

1

Consider: ``` venture[script] > assume x = beta(1, 1) 0.795638860957 venture[script] > observe flip(x) = true [-0.22860988835221455] venture[script] > slice(default, one, 0.2, 10, 1) Warning: skipping annotating did 22, assumed...