Arcade Learning Environment

E95193

benchmark suite reinforcement learning environment collection research platform software framework

Arcade Learning Environment is a widely used research platform that provides a suite of Atari 2600 games for developing and evaluating reinforcement learning algorithms.

Jump to: Surface forms Statements Referenced by

Observed surface forms (1)

Surface form	Occurrences
Atari multi-agent environments	1

Statements (46)

Predicate	Object
instanceOf	benchmark suite ⓘ reinforcement learning environment collection ⓘ research platform ⓘ software framework ⓘ
actionSpace	finite set of joystick and button actions ⓘ
benchmarkLevel	classic control from pixels ⓘ
category	Atari-based RL testbed ⓘ RL benchmark environment ⓘ
domain	artificial intelligence research ⓘ reinforcement learning ⓘ
evaluationMetric	average return over episodes ⓘ game score ⓘ
fullName	Arcade Learning Environment self-link ⓘ
gamePlatform	Atari 2600 ⓘ
goal	enable fair comparison of RL algorithms ⓘ facilitate reproducible RL research ⓘ
implementationLanguage	C++ ⓘ Python bindings ⓘ
includes	multiple Atari 2600 games ⓘ
inputType	raw game screen images ⓘ
inspired	development of standardized RL benchmarks ⓘ
license	open-source license ⓘ
notableUse	benchmarking policy-based RL algorithms ⓘ benchmarking value-based RL algorithms ⓘ evaluation of Deep Q-Networks (DQN) ⓘ evaluation of deep reinforcement learning methods ⓘ
observationSpace	high-dimensional visual observations ⓘ
outputType	game score rewards ⓘ terminal game states ⓘ
platformType	Atari 2600 game platform emulator ⓘ
provides	standardized interface to Atari 2600 games ⓘ suite of Atari 2600 games ⓘ
requires	Atari 2600 emulator ⓘ
researchCommunity	machine learning community ⓘ reinforcement learning community ⓘ
shortName	ALE ⓘ
supports	discrete action spaces ⓘ episodic tasks ⓘ visual observation spaces ⓘ
taskType	single-agent reinforcement learning tasks ⓘ
usedBy	academic researchers ⓘ industry research labs ⓘ
usedFor	benchmarking reinforcement learning agents ⓘ comparing performance of RL algorithms ⓘ developing reinforcement learning algorithms ⓘ evaluating reinforcement learning algorithms ⓘ

Referenced by (4)

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

Atari deep Q-network → environmentFramework → Arcade Learning Environment ⓘ

Arcade Learning Environment → fullName → Arcade Learning Environment self-link ⓘ

PettingZoo → hasEnvironmentType → Arcade Learning Environment ⓘ

this entity surface form: Atari multi-agent environments

OpenAI Gym → influencedBy → Arcade Learning Environment ⓘ