Updated RELEASE-NOTES to include more changes

Author: fonnesbeck

RELEASE-NOTES.md

@@ -1,6 +1,6 @@
 # Release Notes
 
-## PyMC3 3.1 (TBA)
+## PyMC3 3.1 (June 23, 2017)
 
 ### New features
 
@@ -28,16 +28,26 @@
 
 * Much improved variational inference support:
 
-  - [Add Operator Variational Inference (experimental).](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html)
+  - [Added Operator Variational Inference (experimental).](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html)
 
-  - [Add Stein-Variational Gradient Descent as well as Amortized SVGD (experimental).](https://github.com/pymc-devs/pymc3/pull/2183)
+  - [Added Stein-Variational Gradient Descent as well as Amortized SVGD (experimental).](https://github.com/pymc-devs/pymc3/pull/2183)
 
-  - [Add pm.generator() to easily specify mini-batches.](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html#Minibatch-ADVI)
+  - [Added pm.generator() to easily specify mini-batches.](http://pymc-devs.github.io/pymc3/notebooks/bayesian_neural_network_opvi-advi.html#Minibatch-ADVI)
+  
+  - Added full-rank ADVI
 
   - Added various optimizers including ADAM.
+  
+  - Deprecated old ADVI interface
+  
+  - implemented `fit` function as the primary interface to approximation algorithms
+
+* Added support for multidimensional minibatches
 
 * [Sampled posteriors can now be turned into priors for Bayesian updating with a new interpolated distribution.](https://github.com/pymc-devs/pymc3/pull/2163)
 
+* Added `Approximation` class and the ability to convert a sampled trace into an approximation via its `Empirical` subclass.
+
 * `Model` can now be inherited from and act as a base class for user specified models (see pymc3.models.linear).
 
 * Add MvGaussianRandomWalk and MvStudentTRandomWalk distributions.