diff --git a/deep_q_network.py b/deep_q_network.py
index 1294f96..f6a9ee8 100755
--- a/deep_q_network.py
+++ b/deep_q_network.py
@@ -123,7 +123,7 @@ def trainNetwork(s, readout, h_fc1, sess):
             if random.random() <= epsilon:
                 print("----------Random Action----------")
                 action_index = random.randrange(ACTIONS)
-                a_t[random.randrange(ACTIONS)] = 1
+                a_t[action_index] = 1
             else:
                 action_index = np.argmax(readout_t)
                 a_t[action_index] = 1