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 power of convex relaxation: near-optimal matrix completion」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

Emmanuel Candés and I have just uploaded to the arXiv our paper “ The power of convex relaxation: near-optimal matrix completion “, submitted to IEEE Inf. Theory . In this paper we study the matrix completion problem , which one can view as a sort of “non-commutative” analogue of the sparse recovery problem studied in the field of compressed sensing , although there are also some other significant differences between the two problems. The sparse recovery problem seeks to reco

Now we turn to the matrix completion problem. Instead of an unknown vector , we now have an unknown matrix (we use the shorthand here). We will take a specific type of underdetermined linear measurement of M, namely we pick a random subset of the matrix array of some cardinality , and form the random sample of M.

已知结果和反例

Of course, with no further information on M, it is impossible to complete the matrix M from the partial information – we only have pieces of information and need . But suppose we also know that M is low-rank , e.g. has rank less than r; this is an analogue of sparsity, but for matrices rather than vectors. Then, in principle, we have reduced the number of degrees of freedom for M from to something more like , and so (in analogy with compressed sensing) one may now hope to per

This type of problem comes up in several real-world applications, most famously in the Netflix prize . The Netflix prize problem is to be able to predict a very large ratings matrix M, whose rows are the customers, whose columns are the movies, and the entries are the rating that each customer would hypothetically assign to each movie. Of course, not every customer has rented every movie from Netflix, and so only a small fraction of this matrix is actually known. However, if

证明或构造的主线

Actually, one expects to need to oversample the matrix by a logarithm or two in order to have a good chance of exact recovery, if one is sampling randomly. This can be seen even in the rank one case r=1, in which is the product of a column vector and a row vector; let’s consider square matrices for simplicity. Observe that if the sampled coordinates completely miss one of the rows of the matrix, then the corresponding element of u has gone completely unmeasured, and one canno

On the other hand, one cannot hope to complete the matrix if some of the singular vectors of the matrix are extremely sparse. For instance, in the Netflix problem, a singularly idiosyncratic customer (or dually, a singularly unclassifiable movie) may give rise to a row or column of M that has no relation to the rest of the matrix, occupying its own separate component of the singular value decomposition of M; such a row or column is then impossible to complete exactly without

阅读时建议盯住的点

In a recent paper, Candés and Recht proposed solving the matrix completion problem by minimising the nuclear norm (or trace norm )

amongst all matrices consistent with the observed data . This nuclear norm is the non-commutative counterpart to the norm for vectors, and so this algorithm is analogous to the minimisation (or basis pursuit ) algorithm which is effective for compressed sensing (though not the only such algorithm for this task). They showed, roughly speaking, that exact matrix completion (for, say, square matrices for simplicity) is ensured with high probability so long as the singular vector

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