1 二维数据中的查找https://www.nowcoder.com/practice/abc3fe2ce8e146608e868a70efebf62e?tpId=13&tqId=11154&tPage=1&rp=1&ru=/ta/coding-interviews&qru=/ta/coding-interviews/question-ranking 思路 从左下角开始,其右边的元素都比其大,其左边元素都比其小. 这样可以跳过一些点的查找(同理,右上也行 ...

Motivations Model 重写GCN的聚合函数并介绍node-wise sampling 将node-wise sampling泛化为layer-wise sampling 实现方差缩减的layer-wise sampler 采样中的skip-connection来保持二阶相似性 From Node-Wise Sampling to Layer-Wise Sampling符号注明: p相关符号是node-wise, q相关符号是layer-wise. 带$\hat{}$的是相应符号 ...

Representation Learning on Heterogeneous Graph Contributed by Houye Ji. Heterogeneous Graph Embedding Yukuo Cen, Xu Zou, Jianwei Zhang , Hongxia Yang, Jingren Zhou, Jie Tang. Representation Learning for Attributed Multiplex Heterogeneous Network. KDD ...

对于PageRank, $\boldsymbol{\pi}{\mathrm{pr}}=\boldsymbol{A}{\mathrm{rw}} \boldsymbol{\pi}{\mathrm{pr}}, \text { with } \boldsymbol{A}{\mathrm{rw}}=\boldsymbol{A} \boldsymbol{D}^{-1}$ ,$ \boldsymbol{\pi}_{\mathrm{pr}}$中每个元素代表节点的 GCN converges to limit ...

本文对比了常见散度(KL散度，JS散度)和常见距离(尤其是现在很火的Wasserstein距离) F-散度散度是用来衡量两个概率密度p，q区别的函数，即：两个分布的相似程度. $$D _ { f } ( p | q ) = \int q ( x ) f \left( \frac { p ( x ) } { q ( x ) } \right) d x$$这里的$f$需要满足2个条件: $f$是凸函数且$f(1)=0$. 可以证明: 因为$f$是凸函数,由Jensen不等式可知 $E(f(x)) & ...

437. Path Sum III题目:https://leetcode.com/problems/path-sum-iii/ 难度:Easy 思路： https://leetcode.com/problems/path-sum-iii/discuss/141424/Python-step-by-step-walk-through.-Easy-to-understand.-Two-solutions-comparison.-%3A-) 123456789101112131415161718192 ...

198. House Robber难度: Easy 刷题内容 原题连接 https://leetcode.com/problems/house-robber/description/ 内容描述 12345678910111213141516You are a professional robber planning to rob houses along a street. Each house has a certain amount of money stashed, the on ...

难度: Medium 刷题内容 原题连接 https://leetcode.com/problems/3sum/ 内容描述 123456789101112131415Given an array nums of n integers, are there elements a, b, c in nums such that a + b + c = 0? Find all unique triplets in the array which gives the sum of zero.N ...