diff --git a/README.md b/README.md
index e75b14cda025a5b4e92d3d48d33da95473ac4c37..ac15a74b3840a6145451301232441b5c075ed130 100644
--- a/README.md
+++ b/README.md
@@ -58,3 +58,7 @@ When running, the program will read the following weight file. To use the previo
 ```
 weights-learned.hdf5
 ```
+
+## Can DQN play a perfect game?
+This is an interesting question. The following youtube video seems to have the answer. The video explains the design of the system state and demonstrates the AI playing a perfect game with design:
+https://www.youtube.com/watch?v=vhiO4WsHA6c