bluemsi/2nd
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: complete wikification of War Commander batch 1&2 and final grey dot cleanup

Browse Source
This commit is contained in:
yesung
2026-04-27 18:58:22 +09:00
parent 3424166ea2
commit 6b86b0da4c
2706 changed files with 9074 additions and 7273 deletions
Download Patch File Download Diff File Expand all files Collapse all files
10_Wiki/Topics/AI/Proximal-Policy-Optimization.md
+3 -3
View File
@@ -1,12 +1,12 @@
---
id: RL-PPO-001
category: "[[10_Wiki/💡 Topics/AI]]"
category: "10_Wiki/💡 Topics/AI"
confidence_score: 1.0
tags: [ai, reinforcement-learning, ppo, proximal-policy-optimization, openai, stability, policy-gradient]
last_reinforced: 2026-04-26
---
# [[Proximal Policy Optimization (PPO, 근사 정책 최적화)]]
# Proximal Policy Optimization (PPO, 근사 정책 최적화)
## 📌 한 줄 통찰 (The Karpathy Summary)
> "정책의 급격한 변화를 '클리핑(Clipping)'이라는 고삐로 억제하여, 복잡한 환경에서도 무너지지 않는 안정적인 지능의 성장을 견인하라" — OpenAI가 제안한 강화학습 알고리즘으로, 정책 업데이트 폭을 제한함으로써 학습의 안정성과 효율성을 동시에 달성한 현대 RL의 표준 기법.
@@ -25,4 +25,4 @@ last_reinforced: 2026-04-26
## 🔗 지식 연결 (Graph)
- [[Policy-Gradient-Methods]], [[Actor-Critic-Models]], [[Off-policy-vs-On-policy-Learning]], [[Reinforcement-Learning]]
- **Raw Source:** [[10_Wiki/Topics/AI/Proximal-Policy-Optimization.md]]
- **Raw Source:** 10_Wiki/Topics/AI/Proximal-Policy-Optimization.md