Antigravity Agent
34 lines
1.2 KiB
Markdown
34 lines
1.2 KiB
Markdown
---
|
|
id: wiki-2026-0508-non-linear-activation-functions
|
|
title: Non-linear Activation Functions
|
|
category: 10_Wiki/Topics
|
|
status: duplicate
|
|
canonical_id: "[[Leaky-ReLU-and-Activations]]"
|
|
aliases: [Activation Functions, Nonlinearities, ReLU Family]
|
|
duplicate_of: "[[Leaky-ReLU-and-Activations]]"
|
|
source_trust_level: A
|
|
confidence_score: 0.9
|
|
verification_status: redirected
|
|
tags: [redirect, activation, relu, gelu, swish]
|
|
raw_sources: []
|
|
last_reinforced: 2026-05-10
|
|
github_commit: pending
|
|
tech_stack: { language: Python, framework: PyTorch }
|
|
---
|
|
|
|
## REDIRECT → [[Leaky-ReLU-and-Activations]]
|
|
|
|
이 문서는 **[[Leaky-ReLU-and-Activations]]** 로 통합되었다. ReLU/LeakyReLU/PReLU/ELU/GELU/SiLU(Swish)/Mish/SwiGLU 등 비선형 활성화 함수 전반은 정식 문서를 참고하라.
|
|
|
|
## 요약
|
|
- 비선형성이 신경망 표현력의 핵심.
|
|
- 2026 표준: Transformer는 GELU/SwiGLU, CNN은 SiLU/Mish, classic은 ReLU 변형.
|
|
|
|
## 🔗 Graph
|
|
- 부모: [[Leaky-ReLU-and-Activations]] (canonical)
|
|
- 인접: [[Neural Networks]], [[Backpropagation]], [[Vanishing Gradient]]
|
|
|
|
## 🕓 Changelog
|
|
- 2026-05-08 Phase 1 자동 생성
|
|
- 2026-05-10 Manual cleanup — REDIRECT 처리, 정식 문서로 통합
|