Fix typo (#891)

Co-authored-by: holly1238 <[email protected]>

Author: tom-doerr

intermediate_source/reinforcement_q_learning.py

@@ -388,7 +388,7 @@ def plot_durations():
 # single step of the optimization. It first samples a batch, concatenates
 # all the tensors into a single one, computes :math:`Q(s_t, a_t)` and
 # :math:`V(s_{t+1}) = \max_a Q(s_{t+1}, a)`, and combines them into our
-# loss. By defition we set :math:`V(s) = 0` if :math:`s` is a terminal
+# loss. By definition we set :math:`V(s) = 0` if :math:`s` is a terminal
 # state. We also use a target network to compute :math:`V(s_{t+1})` for
 # added stability. The target network has its weights kept frozen most of
 # the time, but is updated with the policy network's weights every so often.