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

[G1-Sync] Manual knowledge update

Browse Source
This commit is contained in:
Antigravity Agent
2026-04-30 22:42:02 +09:00
parent 0bd4f19e38
commit c36c0644a1
4888 changed files with 18470 additions and 18602 deletions
Download Patch File Download Diff File Expand all files Collapse all files
10_Wiki/Topics/Differentiable Programming.md
+3 -3
View File
@@ -1,8 +1,8 @@
---
id: P-REINFORCE-AI-DIF-PROG
id: [[P-Reinforce]]-AI-DIF-PROG
category: "10_Wiki/💡 Topics/AI"
confidence_score: 0.98
tags: [Differentiable Programming, AI, JAX, PyTorch, Optimization]
tags: [Differentiable Programming, AI, JAX, PyTorch, [[Optimization]]]
last_reinforced: 2026-04-20
---
@@ -23,5 +23,5 @@ last_reinforced: 2026-04-20
- 모든 로직을 미분 가능하게 만들기는 어렵다(특히 불연속적인 이산적 선택). 이를 해결하기 위해 `Reinforce` 기법이나 `Gumbel-Softmax` 같은 트릭을 써서 확률적으로 미분 가능하게 연결하는 연구가 활발하다.
## 🔗 지식 연결 (Graph)
- Related: Deep-Reinforcement-Learning , [[Complexity-Theory]]
- Related: Deep-[[Reinforcement-Learning]] , [[Complexity-Theory]]
- Foundation: [[Computational Theory & Math/Information Theory]]