Fix Roeder citation

normal-computing · Nov 5, 2024 · 44b96d6 · 44b96d6
1 parent 6af158b
commit 44b96d6
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/posteriors/vi/dense.py b/posteriors/vi/dense.py
@@ -43,7 +43,7 @@ def build(
         temperature: Temperature to rescale (divide) log_posterior.
         n_samples: Number of samples to use for Monte Carlo estimate.
         stl: Whether to use the stick-the-landing estimator
-            from (Roeder et al](https://arxiv.org/abs/1703.09194).
+            from [Roeder et al](https://arxiv.org/abs/1703.09194).
         init_L: Initial lower triangular matrix $L$ satisfying $LL^T$ = $\\Sigma$.
 
     Returns:

diff --git a/posteriors/vi/diag.py b/posteriors/vi/diag.py
@@ -42,7 +42,7 @@ def build(
         temperature: Temperature to rescale (divide) log_posterior.
         n_samples: Number of samples to use for Monte Carlo estimate.
         stl: Whether to use the stick-the-landing estimator
-            from (Roeder et al](https://arxiv.org/abs/1703.09194).
+            from [Roeder et al](https://arxiv.org/abs/1703.09194).
         init_log_sds: Initial log of the square-root diagonal of the covariance matrix
             of the variational distribution. Can be a tree matching params or scalar.