faceRecgSys:使用Matlab的人脸识别系统; 算法:LBP,PCA,KNN,SVM和朴素贝叶斯
1
Bayesian methods for machine learning have been widely investigated, yielding principled methods for incorporating prior information into inference algorithms. In this survey, we provide an in-depth review of the role of Bayesian methods for the reinforcement learning (RL) paradigm. The major incentives for incorporating Bayesian reasoning in RL are: 1) it provides an elegant approach to action-selection (exploration/ exploitation) as a function of the uncertainty in learning; and 2) it provides a machinery to incorporate prior knowledge into the algorithms. We first discuss models and methods for Bayesian inference in the simple single-step Bandit model. We then review the extensive recent literature on Bayesian methods for model-based RL, where prior information can be expressed on the parameters of the Markov model. We also present Bayesian methods for model-free RL, where priors are expressed over the value function or policy class. The objective of the paper is to provide a comprehensive survey on Bayesian RL algorithms and their theoretical and empirical properties.
2021-11-21 19:28:33 1.81MB 贝叶斯 增强学习 机器学习 深度学习
1
线性高斯分布: P(c | h, subsidy) = N(ath + bt, t2)(c) = 1/ (t21/2) e –1/2{[c-(ath + bt)]/t]} P(c | h, ~subsidy) = N(afh + bf, f2)(c) = 1/ (f21/2) e –1/2{[c-(afh + bf)]/t]} S型函数(Sigmoid function) p(buys | Cost = c) = 1 / {1 + exp[-2(-u+)/ ]}
2021-11-21 16:56:17 1.55MB 贝叶斯
1
贝叶斯分类算法是统计学的一种分类算法,利用概率统计对数据进行分类,含数据集,Python实现贝叶斯分类算法
2021-11-21 15:59:20 1KB 贝叶斯 python 分类算法
1
package subjectivebayes; import java.awt.Toolkit; import javax.swing.SwingUtilities; import javax.swing.UIManager; import java.awt.Dimension; /** * Title: * * Description: * * Copyright: Copyright (c) 2010 * * Company: * * @author not attributable * @version 1.0 */ public class MyApp { boolean packFrame = false; /** * Construct and show the application. */ public MyApp() { EnterBayes frame = new EnterBayes(); // Validate frames that have preset sizes // Pack frames that have useful preferred size info, e.g. from their layout if (packFrame) { frame.pack(); } else { frame.validate(); } // Center the window Dimension screenSize = Toolkit.getDefaultToolkit().getScreenSize(); Dimension frameSize = frame.getSize(); if (frameSize.height > screenSize.height) { frameSize.height = screenSize.height; } if (frameSize.width > screenSize.width) { frameSize.width = screenSize.width; } frame.setLocation((screenSize.width - frameSize.width) / 2, (screenSize.height - frameSize.height) / 2); frame.setVisible(true); } /** * Application entry point. * * @param args String[] */ public static void main(String[] args) { SwingUtilities.invokeLater(new Runnable() { public void run() { try { UIManager.setLookAndFeel(UIManager. getSystemLookAndFeelClassName()); } catch (Exception exception) { exception.printStackTrace(); } new MyApp(); } }); } }
2021-11-21 14:13:14 58KB 主观贝叶斯
1
一提到电子邮件(e.mail),相信大家都不会感到陌生。随着Intemet的迅猛发展,电子邮件凭借使用方便、快捷、廉价的特点很快被广大网络用户所接受,已成为当前最流行的信息交流方式之一。但是电子邮件给我们带来便利的同时,垃圾邮件也随之产生,带来了巨大的危害。近年来大量的商业、色情、反动垃圾邮件和邮件病毒的泛滥给互联网用户带来很多烦恼和侵害,也给社会带来了极大的负面影响,邮件系统的安全问题引起业界的重点关注。垃圾邮件在国内的情况十分严重,中国如今成为了世界垃圾邮件来源的第三大国,反垃圾邮件迫在眉睫。因此研究垃圾邮件过滤具有着极其重大的现实意义。
2021-11-21 10:18:55 2.24MB 反垃圾邮件
1
bayes-python 具体代码见:bayes_iris.py 我直接用了iris_data数据集,每种花我选取前45条数据当做训练集,剩下5条数据另外存入测试集iris_test_data,并将数据随机手动打乱 测试集如下: 因为这个数据集是连续性属性,所以需要利用概率密度函数。 具体实验步骤为: (1)先读取数据集 (2)计算训练数据集上每个类别的各个特征属性上的均值和方差 (3)开始对测试数据集进行分类 (4)首先估计先验概率,这里我每个类别所占整体数据集的比例是一样的 (5)利用概率密度函数,计算测试数据集上各个属性在每个类别上的条件概率 (6)计算后验概率=先验概率*条件概率 (7)比较在各个类别上的后验概率,取最大值,则分为这个类别 结果如下: 我们将结果与测试集比较发现结果完全正确!
2021-11-19 13:44:52 118KB 附件源码 文章源码
1
在matlab开发环境下对贝叶斯网络进行结构学习 模型预测推理分类
2021-11-19 09:34:02 651B k2算法
1
为什么称为最小错误率Bayes决策? 最小错误率Bayes决策使得每个观测值下的条件错误率最小因而保证了(平均)错误率最小。 Bayes决策是最优决策:即,能使决策错误率最小。 为什么Bayes决策是最优决策,我们还要学习这门课的其它方法?
2021-11-18 19:17:02 3.3MB 贝叶斯决策
1
用贝叶斯算法解决数据挖掘中分类的问题,通过训练集先训练出分类规则后,再输入测试集进行测试
2021-11-18 18:43:24 888KB 贝叶斯
1