DDPG

E98481

actor-critic algorithm deep reinforcement learning algorithm model-free reinforcement learning method off-policy reinforcement learning method

DDPG (Deep Deterministic Policy Gradient) is a model-free, off-policy deep reinforcement learning algorithm designed for continuous action spaces, combining ideas from DQN and actor-critic methods.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (3)

Label	Occurrences
DDPG canonical	3
Deep Deterministic Policy Gradient	3
Continuous control with deep reinforcement learning	1

Statements (50)

Predicate	Object
instanceOf	actor-critic algorithm ⓘ deep reinforcement learning algorithm ⓘ model-free reinforcement learning method ⓘ off-policy reinforcement learning method ⓘ
actorObjective	maximize critic-estimated Q-value ⓘ
algorithmFamily	Q-learning inspired methods ⓘ policy gradient methods ⓘ
basedOn	deterministic policy gradient theorem ⓘ
category	continuous-action RL algorithm ⓘ
commonlyEvaluatedOn	MuJoCo benchmarks ⓘ OpenAI Gym continuous control environments ⓘ
commonlyUsedFor	continuous control tasks ⓘ
contrastWith	Atari deep Q-network ⓘ surface form: DQN (which handles discrete actions) stochastic policy gradient methods ⓘ
criticLossType	temporal-difference error ⓘ
criticObjective	minimize Bellman error ⓘ
explorationStrategy	noise added to deterministic policy output ⓘ
fullName	DDPG self-linksurface differs ⓘ surface form: Deep Deterministic Policy Gradient
handlesActionSpaceType	continuous action space ⓘ
inputToActor	state ⓘ
inputToCritic	state-action pair ⓘ
inspiredBy	Atari deep Q-network ⓘ surface form: Deep Q-Network
introducedBy	Alexander Pritzel ⓘ Daan Wierstra ⓘ David Silver ⓘ Jonathan J. Hunt ⓘ Nicolas Heess ⓘ Timothy P. Lillicrap ⓘ Tom Erez ⓘ Yuval Tassa ⓘ
introducedInPaper	DDPG self-linksurface differs ⓘ surface form: Continuous control with deep reinforcement learning
introducedInYear	2015 ⓘ
optimizationMethod	gradient descent ⓘ
outputOfActor	continuous action ⓘ
outputOfCritic	Q-value ⓘ
policyType	deterministic policy ⓘ
stabilityTechnique	experience replay ⓘ target networks ⓘ
trainingParadigm	off-policy learning ⓘ
updateType	bootstrapped TD learning ⓘ
uses	Ornstein–Uhlenbeck process ⓘ surface form: Ornstein-Uhlenbeck noise actor network ⓘ critic network ⓘ experience replay buffer ⓘ exploration noise process ⓘ soft target updates ⓘ target actor network ⓘ target critic network ⓘ
usesFunctionApproximator	deep neural network ⓘ
valueFunctionType	action-value function ⓘ

Referenced by (7)

Full triples — surface form annotated when it differs from this entity's canonical label.

OpenAI Baselines → implementsAlgorithm → DDPG ⓘ

this entity surface form: Deep Deterministic Policy Gradient

TF-Agents → supportsAlgorithmFamily → DDPG ⓘ

Stable Baselines → supportsAlgorithm → DDPG ⓘ

DDPG → fullName → DDPG self-linksurface differs ⓘ

this entity surface form: Deep Deterministic Policy Gradient

DDPG → introducedInPaper → DDPG self-linksurface differs ⓘ

this entity surface form: Continuous control with deep reinforcement learning

Hindsight Experience Replay → compatibleWith → DDPG ⓘ

this entity surface form: Deep Deterministic Policy Gradient