Forgot password
 Register account
View 4|Reply 0

JL引理

[Copy link]

3160

Threads

7930

Posts

48

Reputation

Show all posts

hbghlyj posted 2025-6-22 09:35 |Read mode
Johnson–Lindenstrauss引理表明高维数据可以被投影到低维空间中,同时近似保持点对之间的距离。
我们需要构造一个随机线性映射 $f: \mathbb{R}^N \to \mathbb{R}^n$,使得对于给定的 $\epsilon \in (0,1)$ 和 $m$ 个点 $\{x_1, \ldots, x_m\}$,
\[
(1-\epsilon) \|x_i - x_j\|_2^2 \leq \|f(x_i) - f(x_j)\|_2^2 \leq (1+\epsilon) \|x_i - x_j\|_2^2 \quad \forall \quad 1 \leq i < j \leq m.
\]

Quick Reply

Advanced Mode
B Color Image Link Quote Code Smilies
You have to log in before you can reply Login | Register account

$\LaTeX$ formula tutorial

Mobile version

2025-6-23 08:54 GMT+8

Powered by Discuz!

Processed in 0.014533 seconds, 22 queries