[KongchangAI]
ProductStarCraft

StarCraft

A real-time strategy game developed by Blizzard Entertainment. Owing to its complex strategy space and imperfect-information characteristics, it is used by organizations such as DeepMind as a research platform for multi-agent reinforcement learning.

Timeline (last 90 days)

Nov 1

Multi-agent reinforcement learning has achieved success in game environments such as StarCraft and Dota 2, but these successes often rely on carefully designed reward functions and restricted action spaces.

Unverified50%
Nov 1

The success of multi-agent AI in StarCraft and Dota 2 relies heavily on the centralized training with decentralized execution (CTDE) framework and carefully designed reward shaping.

Unverified50%
Oct 13

TTT draws on the core framework of Werewolf and the StarCraft custom map 'Mafia', with players secretly divided into three basic factions: Innocents, Traitors and Detectives.

Unverified50%
Oct 8

StarCraft simultaneously presents challenges of imperfect information, real-time decision-making, long-horizon planning, and a vast action space, and has long been regarded as a high-difficulty testbed for AI research.

Unverified50%
Sep 12

The original StarCraft builds its science fiction narrative around three races: Terran, Protoss, and Zerg.

Unverified50%
Aug 29

Early 3D-to-2D sprite rendering pipelines were already in use in the 1990s; the sprites in StarCraft and Age of Empires were first modeled in 3D Studio Max and then rendered as 2D bitmaps.

Unverified50%

All Facts (6)

Source Articles