diff --git a/2_reinforcement_learning/RL_tutorial.ipynb b/2_reinforcement_learning/RL_tutorial.ipynb
index ef868b2..c62de6b 100644
--- a/2_reinforcement_learning/RL_tutorial.ipynb
+++ b/2_reinforcement_learning/RL_tutorial.ipynb
@@ -622,7 +622,7 @@
         "\n",
         "Now, if we have a path, we can, for any step t, define the value of the action $a_t$ in the state $s_t$:\n",
         "\n",
-        "$$Q_t(s_t, a_t) = \\sum_{i=0}^{k-t} \\gamma^i r_{t+1} + \\gamma^{k-t+i} V(s_{k+1})$$\n"
+        "$$Q_t(s_t, a_t) = \\sum_{i=0}^{k-t} \\gamma^i r_{t+i+1} + \\gamma^{k-t+1} V(s_{k+1})$$\n"
       ]
     },
     {
diff --git a/2_reinforcement_learning/RL_tutorial_solution.ipynb b/2_reinforcement_learning/RL_tutorial_solution.ipynb
index 4e36fc4..0810127 100644
--- a/2_reinforcement_learning/RL_tutorial_solution.ipynb
+++ b/2_reinforcement_learning/RL_tutorial_solution.ipynb
@@ -685,7 +685,7 @@
         "\n",
         "Now, if we have a path, we can, for any step t, define the value of the action $a_t$ in the state $s_t$:\n",
         "\n",
-        "$$Q_t(s_t, a_t) = \\sum_{i=0}^{k-t} \\gamma^i r_{t+1} + \\gamma^{k-t+i} V(s_{k+1})$$\n"
+        "$$Q_t(s_t, a_t) = \\sum_{i=0}^{k-t} \\gamma^i r_{t+i+1} + \\gamma^{k-t+1} V(s_{k+1})$$\n"
       ]
     },
     {