Lines Matching refs:selected_policy

419     def update_metrics(self, access_time, selected_policy):  argument
426         policy_name = self.policy_names[selected_policy]
1033         selected_policy = max(range(len(self.policies)), key=lambda x: samples[x])
1034         reward = self.policies[selected_policy].generate_reward(key)
1036         self._as[selected_policy] += reward
1037         self._bs[selected_policy] += 1 - reward
1038         return selected_policy
1090         selected_policy = p.argmax()
1091         reward = self.policies[selected_policy].generate_reward(key)
1093         self.A[selected_policy] += np.outer(x_i, x_i)
1094         self.b[selected_policy] += reward * x_i
1095         self.A_inv[selected_policy] = np.linalg.inv(self.A[selected_policy])
1097         return selected_policy