算法
文章平均质量分 52
All about algorithms
tangwing
三千年读史,不外功名利禄,九万里悟道,终归诗酒田园
展开
-
自动驾驶技术的一些关键点
地图众包 专业 Lidar+camera 点云结合定位GPS 北斗 WIFI 基站... Feature定位感知雷达 Lidar(激光) 毫米波 超声波 视觉 detection tracking classification segmentation 预测model based model free车道序列选择规划路径规划 A* 轨迹规划:离散踩点,成本最优,平滑3维:速度,距离,时间控制mpc pi.原创 2021-10-28 10:02:28 · 217 阅读 · 0 评论 -
约束求解应用交流会报告
报告pdf一些个人笔记原创 2021-10-13 10:49:23 · 201 阅读 · 0 评论 -
[NOTE]Discrete Optimization by Prof. Pascal Van Hentenryck
春节假期https://www.coursera.org/learn/discrete-optimization原创 2021-03-04 21:09:52 · 249 阅读 · 0 评论 -
Time Series Forecasting (三) : hierachical and grouped time series
这一篇的主题主要是如何处理一组有层级关系的序列预测。例如我们进行未来30天的游客数量预测,需要涉及多个颗粒度:城市,省和国家。当然我们可以对每个维度单独建模进行预测,但是很明显这多个序列是有层级关系的:城市维度的预测应该可以聚合到省份而进一步向上聚合到国家。各序列单独预测并不能保证这种一致性。那么就涉及到对这一方向的进一步研究。笔者偶然发现Rob J Hyndman的书(见上图)中是有这样一章的Chapter 10Forecasting hierarchical or groupedtime s...原创 2020-12-20 19:23:53 · 362 阅读 · 0 评论 -
Forecasting (二) : Deep Models [ongoing]
Deep modelsDeepARSalinas, D., Flunkert, V., Gasthaus, J., & Januschowski, T. (2020). DeepAR: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting. https://doi.org/10.1016/j.ijforecast.2019.07.001A.原创 2020-12-20 18:39:32 · 352 阅读 · 0 评论 -
Forecasting (一):introduction
本篇文章希望对demand forecasting涉及的技术进行框架性的整理。首先参考的是供应链及库存相关的著作,一般其中都会有关于forecasting的一章。ReferencesWaters, D. (2003).Inventory control and management 2nd. John Wiley & Sons. (偏OM) Axsäter, S. (2015).Inventory control(Vol. 225) 3rd. Springer. (更新,信息更多些)...原创 2020-11-09 10:49:50 · 561 阅读 · 0 评论 -
MLE、MAP、贝叶斯估计、MCMC、EM
机器学习中的MLE、MAP、贝叶斯估计 - 李文哲的文章 - 知乎 https://zhuanlan.zhihu.com/p/37215276上面这篇文章对所提到的三种方法做了清晰的对比。总结图:另外文中总结:几点重要的Take-aways:每一个模型定义了一个假设空间,一般假设空间都包含无穷的可行解; MLE不考虑先验(prior),MAP和贝叶斯估计则考虑先验(prior); MLE、MAP是选择相对最好的一个模型(point estimation), 贝叶斯方法则是通过观测数据来估原创 2020-10-31 14:04:37 · 1418 阅读 · 0 评论 -
[Casual note] Time series prediction
Time series prediction- 在主要feature是时间时用的比较多。如果有很多额外的feature,考虑深度模型- Trend: 长期的趋势- Seasonality:季节性- Cyclicity:周期性 比如经济周期- Irregularity:unpredictable factors。最好在数据层面就剔除掉# Before forecasting- Make sure the series is stationary : devoid of seasonali原创 2020-09-01 10:14:20 · 171 阅读 · 0 评论 -
Multi-agent and distributed RL [Topic under construction]
It's unrealistic to consider the environement as static with only one agent: it's almost always necessary to consider the reaction of multiple agents on each others' moves. With multiple agent, the space of control (action) becomes exponentially large on t原创 2020-08-23 15:50:40 · 165 阅读 · 0 评论 -
Can agents learn inside of their own dreams?
这次阅读一篇NIPS2018的文章,关于World Models in Reinforcement Learning.原创 2020-07-27 16:44:51 · 99 阅读 · 0 评论 -
[NOTE] Advice and Perspectives on RL Research Frontiers - Rich Sutton in DLRLSS 2019
根据我的习惯,当然先放ressources:slides,video. 这是Sutton在DLRLSS 2019 summer school上的一个lecture,从他自己的角度分享了对RL领域的一些理解,他目前的研究方向及前沿等。一些思考还是很有启发的。个别要点摘录于此,细节可以自行阅读、观看。Developing your own research thoughtsThere are no authorities in science. Be ambitious but also humble.原创 2020-07-21 13:48:11 · 154 阅读 · 0 评论 -
[Note In progress] Model-based Reinforcement Learning
Model based methods can be used in Control Theory. Environment has assumptions and approximations.Learn the model. By supervised learning, for instance. Play the game then train the world model. World models: one of my favorite approaches in which th...原创 2020-07-19 14:11:13 · 143 阅读 · 0 评论 -
[NOTE in progress] Simulation Optimization
简单记录一下关于仿真优化的一些知识点和思考。原创 2020-06-14 15:18:20 · 464 阅读 · 0 评论 -
A Road Map for Deep Learning
点这个:https://towardsdatascience.com/a-road-map-for-deep-learning-b9aee0b2919f原创 2020-06-07 10:37:54 · 155 阅读 · 0 评论 -
[NOTE in progress] ECE236C - Optimization Methods for Large-Scale Systems [on going]
Source:http://www.seas.ucla.edu/~vandenbe/ee236c.htmlIntroductionOutlineFirst-order algorithms Decomposition and splitting Second-order algorithms for unconstrained optimization Interior point for conic optimizationGradientConvexity-∇^2(f..原创 2020-06-06 22:57:26 · 477 阅读 · 0 评论 -
【转载】Overview of gradient descent algorithms
Overview of gradient descent algorithmsAn overview of gradient descent optimization algorithmsGradient descent is the preferred way to optimize neural networks and many other machine learning algorithms but is often used as a black box. This post..转载 2020-06-02 07:41:44 · 156 阅读 · 0 评论 -
[NOTE in progress] Distributed Optimization and Statistical Learning via ADMM - Boyd
Reading notes of the paper "Distributed Optimization and Statistical Learning via ADMM" byBoyd, Parikh, Chu, Peleato and Eckstein.IntroductionADMM : developped in the 70s with roots in the 50s. Proved to be highly related to other methods like Douglas.原创 2020-05-17 17:37:23 · 533 阅读 · 0 评论 -
Taylor, Jacobian, Hessian, Newton and all the else about gradient
本文的主要目的是对基于gradient的一些approximation知识点以及优化方法做一个简单的review。详细内容参考引用链接,这里只列出key points,主要是在遗忘的时候能够快速catch up…Jacobian矩阵和Hessian矩阵引用:Jacobian矩阵和Hessian矩阵 雅克比与海森雅可比矩阵(描述的一阶导数矩阵).. 把它理解为一阶gradient就...原创 2020-05-03 11:36:23 · 205 阅读 · 0 评论 -
Benders Decomposition vs Danzig-Wolf Decomposition
本文记录了一些对Benders (B)和Danzig-Wolf(DW) decomposition 的一些初步理解以及两者的使用场景与对比。来源:Jacek Gondzio,https://www.researchgate.net/post/Can_anyone_state_the_difference_or_pros_cons_of_benders_decomposition_vs_...翻译 2020-05-01 21:21:19 · 786 阅读 · 0 评论 -
Stochastic Optimization: Casual Notes
Currently learning stochastic optimization (SO) theory, I will note important content here.Multi-stage vs Multi-periodStage: is defined according to decisions madePeriod: is defined according to t...原创 2020-02-22 08:05:20 · 331 阅读 · 0 评论 -
[Note In progress]Actor-Critic, DDPG and GAN
In the last postOverview of RLwe've seen two different methodologies: Policy Gradient which aims at training a policy (Actor); and Q-Learning which aims at training a state-action value function (Cr...原创 2020-02-16 15:34:18 · 424 阅读 · 0 评论 -
An Casual Overview of Reinforcement Learning
强化学习概览This overview is largely based on this article:https://medium.com/@SmartLabAI/reinforcement-learning-algorithms-an-intuitive-overview-904e2dff5bbc.Model-based vs Model freeModel: world...原创 2020-02-14 22:59:49 · 527 阅读 · 0 评论 -
Activation Functions in Neural Networks
This article is inspired by这里and 这里.- 激活函数的主要意义是为NN加入非线性的元素。在神经学上模仿的是一个神经元是否有效。A Neural Network without Activation function would simply be aLinear regression Model.Neural-Networks are consid...原创 2020-01-31 00:27:41 · 384 阅读 · 0 评论 -
算法思维举要
Everything is optimization. Problem solving = Representation->evaluation->optimization. The solution is continuous or discrete ? If discrete, could brute-force or branching be a choice ? The ...原创 2020-01-29 23:36:49 · 389 阅读 · 2 评论 -
2019新的开始-梳理知识体系和学习路径
转眼已经2019年了,回望过去,经历很多,变化很大,感觉有些恍惚~最大的变化无非几个月之前入职新公司,继续运筹优化方面的算法工作。新公司,新环境,信息量很大。适应了几个月之后,发现有必要重新梳理自己的知识结构,并且用心去整理和沉淀一些东西。思索了半天又搜索了半天,才想到这个已经被我荒废多年的地方…结果竟然登录成功,庆幸CSDN还没有倒闭,以前的文章竟然还在。惊喜之余决定以后就在此处记录我的点滴...原创 2019-01-02 20:46:49 · 440 阅读 · 0 评论 -
C++ 中的异常处理要点
catch 异常时,C++不会通过相关“转换构造函数”来强行匹配,但可用父类匹配子类在catch中尽量使用引用,以免出现“切割”及不必要的拷贝catch(...) 可以捕获一切,{throw;}可以将异常再次抛出没有catch可用时,或异常由析构函数抛出时,系统调用terminate(), 进而调用abort()终止程序。可通过set_terminate()替换terminate函数要确保构造函数原创 2013-03-23 18:32:50 · 992 阅读 · 0 评论 -
循序渐进学unix——上机记录(六),exec
本次上机记录的主题是如何执行外部程序。Unix下有一组名为execxx()的函数:#include unistd.h>extern char **environ;int execl(const char *path, const char *arg, ...);int execlp(const char *file, const char *arg, ...);int ex原创 2012-12-24 16:11:52 · 1428 阅读 · 0 评论 -
利用CMD 批处理 实现自动更改ip
<br />最近由于总需要在教固定ip和动态获取ip这两种网络环境下切换,所以干脆写了个批处理的脚本,自动实现切换功能。不废话了,贴代码:<br /> <br />@echo off:headecho "Usage:Input 0 to change ip setting to auto mode; 1 changes ip setting to wwc mode."set /p option= Please input your choice:if %option% EQU 0 (原创 2010-06-22 09:21:00 · 3476 阅读 · 2 评论 -
清除句中回车的小工具
最近在网上找了基本txt格式的电子书下来看,感觉很不爽,因为回车换行总是出现在句子中间。可能作者是从网页上复制下来的,总之我需要想办法把这些多余的空行删掉。一个小程序如下:package file;import java.io.*;/** * * @author Lane_Shang */public class EnterCleaner {原创 2010-05-02 17:16:00 · 928 阅读 · 0 评论