Files
myWiki/concepts/low-rank-decomposition.md

35 lines
1.3 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
title: "Low-Rank Decomposition: 神经网络低秩压缩"
created: 2026-06-25
updated: 2026-06-25
type: concept
tags: [model-compression, matrix-factorization, efficient-inference]
sources: ["[[sen-mapping-networks]]"]
---
# Low-Rank Decomposition (低秩分解)
Low-Rank Decomposition (LRD) 是神经网络压缩的经典技术:将全连接层的权重矩阵 W ∈ R^{m×n} 近似分解为两个较小矩阵的乘积:
$$W \approx UV^\top$$
其中 U ∈ R^{m×r}V ∈ R^{n×r},且 r ≪ min(m, n)。参数量从 mn → r(m + n)。
## 在 Mapping Networks 中的应用
Mapping Networks 可以用 LRD 进一步降低映射网络的存储需求:不直接生成 W而是生成 U 和 V 两个更小的矩阵,显著减少映射网络的固定权重数量。
## 相关技术
- **SVD 截断**:基于奇异值分解的后训练压缩
- **Pruning**:稀疏化——移除不重要连接
- **Quantization**:降低权重位宽
这三种技术与 Mapping Networks 正交,可以组合使用——如 Table 8 所示Ours* + LRD 和 Ours* + Prune 均保持可用精度。
## 参考
- Eckart & Young, "The approximation of one matrix by another of lower rank", 1936
- Idelbayev & Carreira-Perpiñán, "Low-rank compression of neural nets", CVPR 2020
- [[weight-modulation]]