prettier equations

Author: joaogui1

g3doc/tutorials/framework.md

@@ -17,8 +17,10 @@ Note that the generalized neighbor loss equation is flexible and can have other
 forms besides the one illustrated above. For example, we can also select
 ![equation](http://www.sciweavers.org/tex2img.php?eq=%5C%28%5Csum_%7Bx_j%20%5Cin%20%5Cmathcal%7BN%7D%28x_i%29%7D%5Cmathcal%7BE%7D%28y_i%2Cg_%5Ctheta%28x_j%29%5C%29&bc=White&fc=Black&im=jpg&fs=12&ff=arev&edit=0)
  to be the
-neighbor loss, which calculates the distance between the ground truth \\(y_i\\)
-and the prediction from the neighbor \\(g_\theta(x_j)\\). This is commonly used
+neighbor loss, which calculates the distance between the ground truth
+![equation](http://www.sciweavers.org/tex2img.php?eq=%20%5C%28y_i%5C%29&bc=White&fc=Black&im=jpg&fs=12&ff=arev&edit=0)
+and the prediction from the neighbor ![equation](http://www.sciweavers.org/tex2img.php?eq=%5C%28g_%5Ctheta%28x_j%29%5C%29&bc=White&fc=Black&im=jpg&fs=12&ff=arev&edit=0).
+ This is commonly used
 in adversarial learning
 [(Goodfellow et al., ICLR'15)](https://arxiv.org/pdf/1412.6572.pdf). Therefore,
 NSL generalizes to **Neural Graph Learning** if neighbors are explicitly