ICode9

精准搜索请尝试: 精确搜索
  • Introduction(介绍)2022-05-21 13:00:48

    What is Vue? 什么是Vue Vue(pronounced/vju:/,like view) is a javascript framework for building user interfaces.It builds on top of standard HTML,CSS and JavaScript,and provides a declarative and component-based programming model that helps you efficiently dev

  • Planning to Explore via Self-Supervised World Models2022-05-13 23:00:11

    发表时间:2020(ICML 2020) 文章要点:这篇文章提出了一个Plan2Explore的model based方法,通过self-supervised方法来做Task-agnostic的探索,在这个过程中有效学习了world model,然后可以迁移到下游的具体任务上,实现zero or few-shot RL。具体的,world model包含encoder,dynamics,reward,decod

  • 阅读论文“Polynomial Fitting Algorithm Based on Neural Network” 笔记2022-05-12 18:04:39

      疑问:explicit/implicit polynomial是?区别? 神经网络: three-layer feedforward network-->Taylor series neutral network model   根据泰勒展开,展开点(某一点)的导数可以用作多项式的参数,进而拟合函数 在论文实验中,选择原点展开,taylor coefficient用作神经元的权重        

  • 流体模拟:Position Based Fluid2022-05-12 17:01:37

    目录流体模拟:Position Based Fluid1. 位置动力学(PBD)1.1 算法过程1.2 约束投影步骤1.3 约束投影求解器2 基于位置动力学的流体模拟(PBF)2.1 不可压缩约束和拉格朗日乘子2.2 软约束(Soft constraint)与混合约束力(Constraint Force Mixing, CFM)2.3 拉伸不稳定性2.4 涡轮控制和人工粘性3.

  • 【转载】rocksdb对leveldb做了哪些优化?2022-05-08 21:32:22

    作者:godj链接:https://www.zhihu.com/question/328622742/answer/713388283 RocksDB Features that are not in LevelDB Performance Multithread compaction Multithread memtable inserts Reduced DB mutex holding Optimized level-based compaction style and universal co

  • Deep Learning based Human Pose Estimation using OpenCV-github2022-05-04 18:00:47

    Deep Learning based Human Pose Estimation using OpenCV 1 姿态估计 在本文中,我们将重点关注人体姿态估计,其中需要检测和定位身体的主要部位/关节(例如肩膀、脚踝、膝盖、手腕等)。 1.1 Keypoint Detection Datasets VGG Pose Dataset 单人 25 MPII Human Pose Dataset 多人 1

  • AWS学习笔记(十八):QuickSight, Redshift2022-05-02 20:04:07

    QuickSight Amazon QuickSight允许您的企业中的所有人通过使用自然语言提问以了解您的数据,通过交互式控制面板探索,或自动查找由机器学习支持的模式和异常值。 Redshift Redshift is based on PostgreSQL, but it’s not used for OLTP It's OLAP – online analytical processing

  • 论文阅读-Distantly Supervised Named Entity Recognition via Confidence-Based Multi-Class Positive and Unl2022-04-28 18:34:40

    题目:基于置信度的多类正无标记学习的远程监督命名实体识别 论文地址:https://openreview.net/pdf?id=0gYkM3fk9Bb 源码地址:https://github.com/kangISU/Conf-MPU-DS-NER 摘要:    本文研究了远程监控下的命名实体识别问题。由于外部词典和/或知识库的不完整性,这种远距离注释的训

  • 漏洞利用:Windows本地提权漏洞(CVE-2021-1732)2022-04-28 13:33:03

    0x00简介 2021年2月10日,微软每月的例行补丁包中修复了一个Windows系统本地提权漏洞,本地攻击者可以利用此漏洞提升到system权限,据称此漏洞被用于定向攻击活动。 0x01漏洞概述 该漏洞由函数win32kfull!xxxCreateWi ndowEx 对应用层回调返回数据校验不严导致,本地用户执行漏洞利用程序

  • AutoRec: Autoencoders Meet Collaborative Filtering2022-04-26 16:34:39

    目录概主要内容训练预测疑问代码 Sedhain S., Menon A. K., Sanner S. and Xie L. AutoRec: autoencoders meet collaborative filtering. In International Conference on World Wide Web (WWW), 2015. 概 其实看到推荐系统结合AutoEncoder, 我的第一反应是通过encoder提取特征

  • TransReID: Transformer-based Object Re-Identification [2102.04378v2] - 论文研读系列(3) 个人笔记2022-04-13 09:33:52

    TransReID: Transformer-based Object Re-Identification [2102.04378v2] 论文题目:TransReID: Transformer-based Object Re-Identification 论文地址:http://arxiv.org/abs/2102.04378v2 代码:https://github.com/heshuting555/TransReID 21年2月文章 1、摘要简介 构建了一个基

  • |NO.Z.00037|——————————|BigDataEnd|——|Hadoop&Spark.V11|------------------------------------------|Spa2022-04-12 13:34:27

    [BigDataHadoop:Hadoop&Spark.V11]                                        [BigDataHadoop.Spark内存级快速计算引擎][|章节三|Hadoop|spark|sparkcore:RDD编程高阶&spark原理初探&shuffle原理|]一、Shuffle原理### --- shuffle原理 ~~~ Shuffle的本意是

  • A Framework for Reinforcement Learning and Planning2022-04-12 07:31:06

    发表时间:2020 文章要点:这篇文章是篇综述,主要从RL和planning的异同入手,总结了对解决MDP这类问题的公共因素,放到一起称作framework for reinforcement learning and planning (FRAP)的框架。首先文章提出,RL和planning的主要区别就是环境模型是已知的还是未知的,通常RL用在环境model

  • Transparent Classification with Multilayer Logical Perceptrons and Random Binarization2022-04-08 17:32:33

    目录概主要内容输入简化 Wang Z., Zhang W, Liu N. and Wang J. Transparent classification with multilayer logical perceptrons and random binarization. In AAAI Conference on Artificial Intelligence (AAAI), 2020. 概 和这儿类似的rule-based的网络, 主要探讨如何训练

  • Forward-Backward Reinforcement Learning2022-04-03 13:34:05

    发表时间:2018 文章要点:这篇文章提出了Forward-Backward Reinforcement Learning (FBRL)算法,在假设reward function和goal已知的情况下,将model free的forward step和model based的backward step结合起来,提升训练效果。 具体的,训练一个反向模型,用\(s_{t+1},a_t\)预测\(s_t\),具体做

  • You are carrying out an investigation to determine how your pet Basilisk Lizard is able to "wal2022-03-31 18:32:04

    You are carrying out an investigation to determine how your pet Basilisk Lizard is able to "walk" across the surface of a liquid. The two liquids you are investigating are water and oil. The water is H2O and the oil is made up of really long-cha

  • 【CVPR 2022】 论文阅读:Ray3D: ray-based 3D human pose estimation for monocular absolute 3D localization2022-03-29 11:34:49

    论文地址:https://arxiv.org/abs/2203.11471 Github:https://github.com/YxZhxn/Ray3D 单位:Aibee、北京工商大学   摘要 在本文中,我们提出了一种新的基于单目光线的3D(Ray3D)绝对人体姿势估计方法。从单目二维位姿输入精确且可推广的绝对三维人体位姿估计是一个不适定问题。为了解决

  • 【学生心理学优化算法】基于学生心理学优化算法求解单目标优化问题(SPBO)含Matlab源码2022-03-21 19:02:33

    1 简介 学生心理学优化算法(Student psychology based optimization algorithm,SPBO)是于2020 提出的一种基于学生向往提高成绩的心理提出的新颖智能优化算法。​ In this article, a new metaheuristic optimization algorithm (named as, student psychology based optimization

  • 论文阅读——A Comprehensive Study on Deep Learning-Based 3D Hand Pose Estimation Methods综述阅读22022-03-21 19:00:04

    3D手势姿态估计综述     本文通过对大量有代表性的论文研究,提出一种基于输入数据模式的新分类法,即RGB、深度或多模态信息。最后,我们展示了在最流行的RGB和基于深度的数据集上的结果,并讨论了这一快速增长领域的潜在研究方向。   1. Introduction 在实施无标记手姿势估计方

  • Codeforces Round #778 (Div. 1 + Div. 2, based on Technocup 2022 Final Round)2022-03-21 10:33:42

    Codeforces Round #778 (Div. 1 + Div. 2, based on Technocup 2022 Final Round) A 题意 进行一次翻转,求相邻两数最大和 思路 总可以将最大的两个凑一起 #include<bits/stdc++.h> #define yes puts("yes"); #define inf 0x3f3f3f3f #define ll long long #define linf 0x3f3f3f3f

  • 读程序写结果_22022-03-20 15:03:27

    源程序: #include <iostream>using namespace std; class Based{public: Based() { cout<<"Based构造函数\n"; fun(); } virtual void fun() { cout<<"Base::fun()函数\n"; }};class Derived:public Based{public: Derived() { cout<&l

  • TEMPORAL DIFFERENCE MODELS: MODEL-FREE DEEP RL FOR MODEL-BASED CONTROL2022-03-19 12:37:17

    发表时间:2018(ICLR 2018) 文章要点:这篇文提出了temporal difference models(TDMs)算法,把goal-conditioned value functions和dynamics model联系起来,建立了model-free和model-based RL的关系,结合了各自的优点,既利用dynamics里丰富的信息,也超过了直接model based RL的效果。 具体的,

  • XML-Based Injection in Spring 基于XML配置的依赖注入2022-03-19 10:06:26

    原文链接:XML-Based Injection in Spring  1. Introduction 概述 In this basic tutorial, we’ll learn how to do simple XML-based bean configuration with the Spring Framework. 这里,一起学习如何在Spring框架中使用XML文件进行bean对象配置吧   2. Overview 回顾一下 Le

  • [转]Jira ScriptRunner Update Priority based on a Custom Field2022-03-19 08:03:31

    本文转自:https://library.adaptavist.com/entity/update-priority-based-on-a-custom-field Overview This script is used as a post function to change the priority of an issue based on a single-select field. Example As a product manager, I have many issues with a

  • Evaluation of Machine Learning Algorithms in Network-Based Intrusion Detection System2022-03-11 12:58:46

    本文提出了一种更好的对测试集进行有效评估的方法,从而防止训练造成过拟合现象。经过实验表明,SVM和ANN对过拟合的免疫能力是最强的。链接为:https://arxiv.org/abs/2203.05232。 Cybersecurity has become one of the focuses of organisations. The number of cyberattacks ke

专注分享技术,共同学习,共同进步。侵权联系[81616952@qq.com]

Copyright (C)ICode9.com, All Rights Reserved.

ICode9版权所有