20260706:新增一些文章
This commit is contained in:
36
concepts/loss-landscape.md
Normal file
36
concepts/loss-landscape.md
Normal file
@@ -0,0 +1,36 @@
|
||||
---
|
||||
title: "Loss Landscape: 神经网络的损失景观"
|
||||
created: 2026-06-25
|
||||
updated: 2026-06-25
|
||||
type: concept
|
||||
tags: [optimization, deep-learning-theory, loss-surface, mode-connectivity]
|
||||
sources: ["[[sen-mapping-networks]]"]
|
||||
---
|
||||
|
||||
# Loss Landscape (损失景观)
|
||||
|
||||
Loss Landscape 指将神经网络训练目标 L(θ) 视为参数空间 R^P 上的高维曲面时的几何结构。该视角为理解泛化、优化难度和参数空间结构提供了关键洞察。
|
||||
|
||||
## 关键发现
|
||||
|
||||
### 低维结构
|
||||
- **Intrinsic Dimension**:深度网络的 objective landscape 的有效内在维度远低于 P(Li et al., 2018)
|
||||
- 训练轨迹聚焦在远低于参数空间维度的子空间上
|
||||
|
||||
### Mode Connectivity
|
||||
- SGD 找到的不同局部极小值之间存在**低损路径**(Garipov et al., 2018)
|
||||
- 表明有效解分布在连通区域而非孤立点
|
||||
|
||||
### 平坦 vs 尖锐极小值
|
||||
- **平坦极小值**:泛化更好(Keskar et al., 2017)
|
||||
- Entropy-SGD 等方法显式偏置梯度下降走向宽阔山谷(Chaudhari et al., 2019)
|
||||
|
||||
## 对 Mapping Networks 的意义
|
||||
|
||||
Mapping Networks 通过**结构性约束**将搜索空间限制在低维流形上,自然地偏向发现更平坦、更鲁棒的参数解——这是一种通过架构选择实现隐式正则化的方式。
|
||||
|
||||
## 参考
|
||||
|
||||
- [[intrinsic-dimension]]
|
||||
- [[manifold-hypothesis]]
|
||||
- [[weight-manifold-hypothesis]]
|
||||
Reference in New Issue
Block a user