Unverified50% confidenceFactExact time
MAPPO is a multi-agent extension of PPO that uses a shared policy network and a centralized value function; it performs well on benchmarks such as StarCraft and has good training stability.
1
Sources
50%
Confidence
Long-term
Relevance
9/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMAPPO采用集中训练、分散执行（CTDE）范式，在StarCraft II单位微操、多机器人仓储等场景中被证明有效72% similarVerifiedPPO（近端策略优化）以其训练稳定性成为连续控制任务的事实标准，SAC（软演员-评论家）通过最大熵框架在样本效率上具有优势68% similarVerifiedPPO（近端策略优化）通过限制每次参数更新的幅度，在学习效率与训练稳定性之间取得平衡67% similarUnverifiedPPO has become one of the most popular RL algorithms in game AI and robot control due to its training stability and implementation simplicity.66% similarVerifiedPPO prevents training collapse caused by excessively large policy updates by introducing trust region constraints, and is one of the most widely applied foundational algorithms in RL training of large models today.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/890311API
curl https://kongchang.com/api/v1/knowledge/claims/890311MCP
get_claim(id=890311)