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

docs: finalized wiki integrity maintenance (v3.0 standard) - pruned 1400+ stubs and fixed 11k+ ghost links

Browse Source
This commit is contained in:
Antigravity Agent
2026-05-02 09:18:34 +09:00
parent c84dcb8371
commit 6445fcc05b
13150 changed files with 55394 additions and 100862 deletions
Download Patch File Download Diff File Expand all files Collapse all files
10_Wiki/Topics_Blog/Loss Functions.md
+2 -2
View File
@@ -6,7 +6,7 @@ tags: [auto-reinforced, loss-functions, optimization, machine-learning, error-me
last_reinforced: 2026-04-20
---
# [[Loss Functions]]
# [[Loss Functions|Loss Functions]]
## 📌 한 줄 통찰 (The Karpathy Summary)
> "모델의 반성문 도우미: AI의 예측이 실제 정답과 얼마나 동떨어져 있는지 수치(Penalty)로 계산하여, 모델이 스스로 '아, 내가 이만큼 틀렸구나'를 깨닫고 정답을 향해 가중치를 수정하게 만드는 성적표이자 학습의 이정표."
@@ -25,6 +25,6 @@ last_reinforced: 2026-04-20
- **정책 변화(RL Update)**: 단순히 오차를 줄이는 정책을 넘어, 모델이 너무 복잡해지는 것을 막기 위해 '벌칙항(Penalty term)'을 손실 함수에 추가하여 일반화 성능을 높이는 정규화 정책이 필수화됨. (L2-Regularization와 연결)
## 🔗 지식 연결 (Graph)
- [[Gradient-Descent]], [[Optimization]], [[Information-Entropy]], [[L2-Regularization]], [[DPO (Direct Preference Optimization)]]
- [[Gradient-Descent|Gradient-Descent]], [[Optimization|Optimization]], [[Information-Entropy|Information-Entropy]], [[L2-Regularization|L2-Regularization]], [[DPO (Direct Preference Optimization)|DPO (Direct Preference Optimization)]]
- **Modern Tech/Tools**: KL Divergence, Huber Loss, Hinge Loss, Log Loss.
---