Add assessment

assessment.md

@@ -0,0 +1,17 @@
+# Tic Tac Toe Lab Assessment
+
+## Checkpoint 1
+Looks good. 
+
+## Checkpoint 2
+Nice! The ranges make use of the fact that there are some useful patterns 
+in the indices of row- and colum-based wins.
+
+## Checkpoint 3
+> The current and future reward for this state is 0 which tells me there are as many
+> games where x wins as loses and most games end in a draw? Does this mean it's a
+> fair game?
+
+Almost--but this isn't a propabilistic process. Having the initial state's
+current and future reward be zero means that if everyone plays perfectly, 
+the game will always be a draw.