html/table_5cf04dbc.html

<table><tr><td/><td>Actor Critic (AC) is a reinforcement learning algorithm in the Policy Gradient method that introduces an additional neural network, allowing two networks to be updated simultaneously. One network approximates the policy, while the other serves as a replacement for the Q function.</td></tr></table>