Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Actor-critic algorithm
open-in-new
<p>The actor-critic algorithm (AC) is a family of <a href="/facts/Reinforcement_learning/NrgPPS0Q">reinforcement learning</a> (RL) algorithms that combine policy-based RL algorithms such as <a href="/facts/Policy_gradient_method/NrgPPS0Q">policy gradient methods</a>, and value-based RL algorithms such as value iteration, <a href="/facts/Q-learning/emySKrgI">Q-learning</a>, <a href="/facts/State%25E2%2580%2593action%25E2%2580%2593reward%25E2%2580%2593state%25E2%2580%2593action/kmv4vSKG">SARSA</a>, and <a href="/facts/Temporal_difference_learning/FJDMDwBD">TD learning</a>. </p><p>An AC algorithm consists of two main components: an "actor" that determines which actions to take according to a policy function, and a "critic" that evaluates those actions according to a value function. Some AC algorithms are on-policy, some are off-policy. Some apply to either continuous or discrete action spaces. Some work in both cases. </p>