def update_q(self, q, new_state: GameState, previous_state: GameState, action_taken: tuple, reward: int) -> dict:
"""
Update q based on:
q[s,a] = Q[s,a] + alpha(r + gamma* max_args(Q[s'])
The main update function for updating a given q value.
"""
previous_state_hash = previous_state.value()
if action_taken is None:
action_taken = ("end", "end")
if previous_state_hash not in q:
q[previous_state_hash] = {}
if action_taken not in q[previous_state_hash]:
q[previous_state_hash][action_taken] = reward
current_q_value = q[previous_state_hash][action_taken]
q[previous_state_hash][action_taken] = \
current_q_value + self.alpha * (reward + (self.gamma * self.get_max_value(new_state.value(), q)) - current_q_value)
return q
