diff --git a/reinforcement_learning/actor_critic.py b/reinforcement_learning/actor_critic.py
index c5a3ee6d79..b58437683e 100644
--- a/reinforcement_learning/actor_critic.py
+++ b/reinforcement_learning/actor_critic.py
@@ -141,7 +141,7 @@ def main():
     for i_episode in count(1):
 
         # reset environment and episode reward
-        state, _ = env.reset()
+        state = env.reset()
         ep_reward = 0
 
         # for each episode, only run 9999 steps so that we don't
@@ -152,7 +152,7 @@ def main():
             action = select_action(state)
 
             # take the action
-            state, reward, done, _, _ = env.step(action)
+            state, reward, done, _ = env.step(action)
 
             if args.render:
                 env.render()
diff --git a/reinforcement_learning/reinforce.py b/reinforcement_learning/reinforce.py
index 961598174c..8c1cda5d8a 100644
--- a/reinforcement_learning/reinforce.py
+++ b/reinforcement_learning/reinforce.py
@@ -81,11 +81,11 @@ def finish_episode():
 def main():
     running_reward = 10
     for i_episode in count(1):
-        state, _ = env.reset()
+        state = env.reset()
         ep_reward = 0
         for t in range(1, 10000):  # Don't infinite loop while learning
             action = select_action(state)
-            state, reward, done, _, _ = env.step(action)
+            state, reward, done, _ = env.step(action)
             if args.render:
                 env.render()
             policy.rewards.append(reward)