-
经典TensorFlow LSTM股票预测实战 真实美股数据落地教程
基于TensorFlow搭建LSTM循环神经网络，可高效完成股票时序价格预测，相较于传统仿真数据训练模型，采用标普500真实历史数据训练的模型适…
-
读懂「Google Wave」：先把定义、反例和适用范围钉死
As readers of this blog are no doubt aware, I (in conjunction with Tim G…
-
「Random matrices: universality of local eigenvalu」：数学直觉如何落到可检验的步骤
Van Vu and I have just uploaded to the arXiv our paper “Random matrices:…
-
「Talagrand’s concentration inequality」：数学直觉如何落到可检验的步骤
In the theory of discrete random matrices (e.g. matrices whose entries a…
-
2026大模型毒性治理全流程：从检测攻防到文本脱敏落地实操
大语言模型依托海量互联网文本预训练，必然会习得网络文本中存在的辱骂、歧视、偏见等毒性内容，极易在生成内容中输出有害文本。当前大模型毒性治理的核心…
-
把「The Szemerédi-Trotter theorem and the cell decom」写成可执行的阅读路径
The celebrated Szemerédi-Trotter theorem gives a bound for the set of in…
-
读懂「DHJ: Still writing the second paper」：先把定义、反例和适用范围钉死
This is a continuation of the previous thread here in the polymath1 proj…
-
读懂「Global regularity of wave maps VI. Abstract theo」：先把定义、反例和适用范围钉死
I’ve just uploaded to the arXiv my paper “Global regularity of wave maps…
-
「Global regularity for a logarithmically supercri」笔记：问题从哪来、证明卡在哪
I’ve just uploaded to the arXiv my paper “Global regularity for a logari…
-
读懂「An equivalence between inverse sumset theorems a」：先把定义、反例和适用范围钉死
For a number of reasons, including the start of the summer break for me …