Kiran1510
diff --git a/‎README.md‎
Lines changed: 1 addition & 1 deletion b/‎README.md‎
Lines changed: 1 addition & 1 deletion
diff --git a/‎figures/comparison_by_representation.png‎
2.2 KB b/‎figures/comparison_by_representation.png‎
2.2 KB
diff --git a/‎figures/comparison_compact.png‎
1.95 KB b/‎figures/comparison_compact.png‎
1.95 KB
diff --git a/‎figures/comparison_double_dqn.png‎
185 KB b/‎figures/comparison_double_dqn.png‎
185 KB
diff --git a/‎figures/comparison_extended.png‎
1.64 KB b/‎figures/comparison_extended.png‎
1.64 KB
diff --git a/‎figures/comparison_local.png‎
1.73 KB b/‎figures/comparison_local.png‎
1.73 KB
diff --git a/‎figures/comparison_mlp_sarsa.png‎
-183 KB b/‎figures/comparison_mlp_sarsa.png‎
-183 KB
diff --git a/‎figures/final_performance.png‎
1.9 KB b/‎figures/final_performance.png‎
1.9 KB
diff --git a/‎figures/heatmap.png‎
2.27 KB b/‎figures/heatmap.png‎
2.27 KB
diff --git a/‎…es/learning_curve_mlp_sarsa__compact.png‎ ‎…s/learning_curve_double_dqn__compact.png‎figures/learning_curve_mlp_sarsa__compact.png renamed to figures/learning_curve_double_dqn__compact.png b/‎…es/learning_curve_mlp_sarsa__compact.png‎ ‎…s/learning_curve_double_dqn__compact.png‎figures/learning_curve_mlp_sarsa__compact.png renamed to figures/learning_curve_double_dqn__compact.png
@@ -25,7 +25,7 @@ Most existing Snake RL work uses off-policy DQN with a single state representati
 
 ### The best agent in action
 
-![Double DQN + Extended representation](recordings/mlp_sarsa__extended.gif)
+![Double DQN + Extended representation](recordings/double_dqn__extended.gif)
 
 *Double DQN with experience replay on the Extended (126d) representation — the strongest configuration (mean score 29.05, max 66).*