qzmjef 发表于 2024-7-1 08:54:16

浅谈人工智能怎么入门


    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">想要</span>入门人工智能,<span style="color: black;">必须</span>两个<span style="color: black;">工具</span>,即算法知识和<span style="color: black;">海量</span>的数据,外加一台计算机,<span style="color: black;">倘若</span>有GPU就更好了。</p>
    <div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic1.zhimg.com/80/v2-8398550639e323f78660ba269c63b864_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">1、学习并<span style="color: black;">把握</span><span style="color: black;">有些</span>数学知识</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">高等数学是<span style="color: black;">基本</span>中的<span style="color: black;">基本</span>,一切理工科都<span style="color: black;">必须</span>这个打底,数据挖掘、人工智能、模式识别此类跟数据打交道的又尤其<span style="color: black;">必须</span>多元微积分运算<span style="color: black;">基本</span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">线性代数很<span style="color: black;">要紧</span>,<span style="color: black;">通常</span><span style="color: black;">来讲</span>线性模型是你最先要<span style="color: black;">思虑</span>的模型,加上很可能要处理多维数据,你<span style="color: black;">必须</span>用线性代数来简洁清晰的描述问题,为分析求解奠定<span style="color: black;">基本</span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">概率论、数理统计、随机过程<span style="color: black;">更加是</span>少不了,<span style="color: black;">触及</span>数据的问题,不确定性几乎是不可避免的,引入随机变量顺理成章,<span style="color: black;">关联</span>理论、<span style="color: black;">办法</span>、模型非常丰富。<span style="color: black;">非常多</span><span style="color: black;">设备</span>学习的算法都是<span style="color: black;">创立</span>在概率论和统计学的<span style="color: black;">基本</span>上的,<span style="color: black;">例如</span>贝叶斯<span style="color: black;">归类</span>器、高斯隐马尔可夫链。</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">再<span style="color: black;">便是</span>优化理论与算法,除非你的问题是像二元一次方程求根那样有现成的公式,否则你将不得不面对<span style="color: black;">各样</span>看起来<span style="color: black;">没</span>解<span style="color: black;">然则</span>要解的问题,优化将是你的GPS为你指路</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">有以上这些知识打底,就<span style="color: black;">能够</span>开拔了,针对<span style="color: black;">详细</span>应用再<span style="color: black;">弥补</span><span style="color: black;">关联</span>的知识与理论,<span style="color: black;">例如</span>说<span style="color: black;">有些</span>我觉得有<span style="color: black;">帮忙</span>的是数值计算、图论、拓扑,更理论一点的还有实/复分析、测度论,偏工程类一点的还有信号处理、数据结构。</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">2、<span style="color: black;">把握</span>经典<span style="color: black;">设备</span>学习理论和算法</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">倘若</span>有时间<span style="color: black;">能够</span>为自己<span style="color: black;">创立</span>一个<span style="color: black;">设备</span>学习的知识图谱,并争取<span style="color: black;">把握</span>每一个经典的<span style="color: black;">设备</span>学习理论和算法,我简单地总结如下:</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">①回归算法:<span style="color: black;">平常</span>的回归算法<span style="color: black;">包含</span>最小二乘法(OrdinaryLeast Square),<span style="color: black;">规律</span>回归(Logistic Regression),逐步式回归(Stepwise Regression),多元自适应回归样条(MultivariateAdaptive Regression Splines)以及本地散点平滑估计(Locally Estimated Scatterplot Smoothing);</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">② 基于实例的算法:<span style="color: black;">平常</span>的算法<span style="color: black;">包含</span> k-Nearest Neighbor(KNN), 学习矢量量化(Learning Vector Quantization, LVQ),以及自组织映射算法(Self-Organizing Map , SOM);</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">③ 基于正则化<span style="color: black;">办法</span>:<span style="color: black;">平常</span>的算法<span style="color: black;">包含</span>:Ridge Regression, Least Absolute Shrinkage and Selection Operator(LASSO),以及弹性网络(Elastic Net);</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">④决策树学习:<span style="color: black;">平常</span>的算法<span style="color: black;">包含</span>:<span style="color: black;">归类</span>及回归树(ClassificationAnd Regression Tree, CART),ID3 (Iterative Dichotomiser 3),C4.5, Chi-squared Automatic Interaction Detection(CHAID), Decision Stump, 随机森林(Random Forest),多元自适应回归样条(MARS)以及梯度推进机(Gradient Boosting Machine,GBM);</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">⑤ 基于贝叶斯<span style="color: black;">办法</span>:<span style="color: black;">平常</span>算法<span style="color: black;">包含</span>:朴素贝叶斯算法,平均单依赖估计(AveragedOne-Dependence Estimators, AODE),以及Bayesian Belief Network(BBN);</p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">⑥ 基于核的算法:<span style="color: black;">平常</span>的算法<span style="color: black;">包含</span>支持向量机(SupportVector Machine</p>




nykek5i 发表于 2024-8-23 02:15:35

顶楼主,说得太好了!

流星的美 发表于 2024-9-8 04:50:53

感谢你的精彩评论,为我的思绪打开了新的窗口。

4zhvml8 发表于 2024-10-7 06:39:17

软文发布平台 http://www.fok120.com/

j8typz 发表于 2024-10-30 12:10:30

谷歌外链发布 http://www.fok120.com/

b1gc8v 发表于 2024-11-12 21:13:11

i免费外链发布平台 http://www.fok120.com/
页: [1]
查看完整版本: 浅谈人工智能怎么入门