陶哲轩博客写数学问题时，通常先把对象定义清楚，再给直觉、反例和证明轮廓。把「The strong law of large numbers」改写成可阅读的中文笔记，重点是：问题在问什么、已知到哪一步、下一步最容易走偏在哪。原站广告、分享条和导航已去掉。

问题在问什么

Let X be a real-valued random variable, and let be an infinite sequence of independent and identically distributed copies of X. Let be the empirical averages of this sequence. A fundamental theorem in probability theory is the law of large numbers , which comes in both a weak and a strong form:

Weak law of large numbers. Suppose that the first moment of X is finite. Then converges in probability to , thus for every .

已知结果和反例

Strong law of large numbers. Suppose that the first moment of X is finite. Then converges almost surely to , thus .

[The concepts of convergence in probability and almost sure convergence in probability theory are specialisations of the concepts of convergence in measure and pointwise convergence almost everywhere in measure theory.]

证明或构造的主线

(If one strengthens the first moment assumption to that of finiteness of the second moment , then we of course have a more precise statement than the (weak) law of large numbers, namely the central limit theorem , but I will not discuss that theorem here. With even more hypotheses on X, one similarly has more precise versions of the strong law of large numbers, such as the Chernoff inequality , which I will again not discuss here.)

The weak law is easy to prove, but the strong law (which of course implies the weak law, by Egoroff’s theorem ) is more subtle, and in fact the proof of this law (assuming just finiteness of the first moment) usually only appears in advanced graduate texts. So 一个常见想法是 I would present a proof here of both laws, which proceeds by the standard techniques of the moment method and truncation. The emphasis in this exposition will be on motivation and methods rather than brevity and

阅读时建议盯住的点

The moment method seeks to control the tail probabilities of a random variable (i.e. the probability that it fluctuates far from its mean) by means of moments, and in particular the zeroth, first or second moment. The reason that this method is so effective is because the first few moments can often be computed rather precisely. The first moment method usually employs Markov’s inequality

(which follows by taking expectations of the pointwise inequality ), whereas the second moment method employs some version of Chebyshev’s inequality , such as

阅读和落地时建议先做的 5 件事

1. 用自己的语言重写定义和结论，不看原文能不能说清对象是什么。
2. 找一个最小反例或边界情形，确认假设少一条会怎样。
3. 把证明拆成可独立检验的引理，每步只保留一个新想法。
4. 若涉及计算或形式化，先写可复现的小例子，再谈一般情形。
5. 记下尚未解决的缺口：缺估计、缺构造，还是缺正确的范畴。

和智能体、形式化工具怎么接

龙虾PRO做 OpenClaw 落地时，数学笔记最有用的部分往往是「可检验的步骤」：定义、反例、引理边界。智能体适合帮忙展开计算和检索，不适合代替你决定哪条假设能扔。

本文侧重全链路风控方法论。落地时请用自身业务单据做回放验证，不要把示例阈值直接当生产策略。 相关：风控体检 · 方案资源