当前位置:网站首页>Online and offline problems
Online and offline problems
2022-07-06 06:12:00 【Zhan Miao】
1. Online questions
Not knowing all the instance information when making decisions , Decisions made cannot be changed after more information is presented .
2. Offline problems
Examples of all known problems before decision .
3. Online algorithms
You can process the input one by one in a serialized way , In other words, you don't need to know all the inputs at the beginning . Relative , For one Offline algorithm , Know all the data you need to enter at the beginning of the problem , And output the results immediately after solving a problem . for example , Selection sort Before sorting, you need to know all the elements to be sorted , However Insertion sort You don't have to .
Because the online algorithm does not know the whole input , So the choices it is forced to make may eventually prove to be suboptimal , The research on online algorithm mainly focuses on how to make a choice in the current environment . Online algorithms for the same problem and Offline algorithm The above viewpoint is formed by the comparative analysis of . If you want to know about online algorithms from other perspectives, you can take a look Stream Algorithm ( Focus on the amount of memory used to accurately present past inputs ), Dynamic algorithm ( Focus on what is needed to maintain an online input result Time complexity ) And online machine learning .
A good example of the concept of online algorithms is the Canadian Traveler Problem , The goal of this problem is to reach a target node with the least cost in a weighted graph , But some edges of this weighted graph are unreliable , May have been eliminated . However, a traveler can only determine whether an edge has been removed when he reaches an endpoint of that edge . In the worst case , The problem will become simple , That is, all uncertain edges are removed, and the problem will become common Shortest path problem .
4. Offline algorithm
Offline algorithm design strategies are based on the basic assumption that the input data is known before the implementation of the algorithm , in other words , For an offline algorithm , Know all the data you need to enter at the beginning of the problem , And output the results immediately after solving a problem , Usually, this kind of algorithm designed under the premise of complete information of the problem is called offline Algorithm .
reference
The concept of online algorithm and offline Algorithm - daiyl0320 - Blog Garden
边栏推荐
猜你喜欢
![[wechat applet] build a development tool environment](/img/f6/51f97b1c927337b34c5b3a4207abb4.png)
[wechat applet] build a development tool environment

Configuring OSPF GR features for Huawei devices

HCIA复习

Manhattan distance sum - print diamond
![[paper reading] nflowjs: synthetic negative data intensive anomaly detection based on robust learning](/img/9c/2753f68ecec3555aaca23800dada1e.png)
[paper reading] nflowjs: synthetic negative data intensive anomaly detection based on robust learning
![[Thesis code] SML part code reading](/img/3c/0deccf499d9b1cbe30a302cb115d73.png)
[Thesis code] SML part code reading

【Postman】测试(Tests)脚本编写和断言详解

How Huawei routers configure static routes

数学三大核心领域概述:代数

异常检测方法总结
随机推荐
LeetCode 729. 我的日程安排表 I
[postman] dynamic variable (also known as mock function)
【Postman】动态变量(也称Mock函数)
MySQL之数据类型
【微信小程序】搭建开发工具环境
selenium源码通读·9 |DesiredCapabilities类分析
功能安全之故障(fault),错误(error),失效(failure)
IPv6 comprehensive experiment
请求转发与重定向
Network protocol model
Pat (Grade B) 2022 summer exam
GTSAM中李群的运用
Eigen sparse matrix operation
Arrays and collections
[C language] qsort function
公司視頻加速播放
【Postman】测试(Tests)脚本编写和断言详解
[postman] test script writing and assertion details
Thoughts on data security (Reprint)
Gtest之TEST宏的用法