Introduction to Restricted Boltzmann Machines

机器学习之RBM(Restricted Boltzmann Machines,受限玻尔兹曼机) 本人最近在学习DBN(Deep Belief Net,深度信念网络),通过学习才知道有RBM这个东西。因为我所要用到的DBN是有RBM通过累加堆叠组成的,要学习DBN就要弄明白RBM的原理。我就在此说一下我自己对RBM的认识和了解,同时也希望对别人有些帮助。 所谓受限玻尔兹曼机就是对玻尔兹曼机进行简化,使玻尔兹曼机更容易更简单地使用,原本玻尔兹曼机的隐元和显元之间是全连接的,而且隐元和隐元之间也是 阅读详情

Introduction

Suppose you ask a bunch of users to rate a set of movies on a 0-100 scale. In classical factor analysis, you could then try to explain each movie and user in terms of a set of latent factors. For example, movies like Star Wars and Lord of the Rings might have strong associations with a latent science fiction and fantasy factor, and users who like Wall-E and Toy Story might have strong associations with a latent Pixar factor.

Restricted Boltzmann Machines essentially perform a binary version of factor analysis. (This is one way of thinking about RBMs; there are, of course, others, and lots of different ways to use RBMs, but I’ll adopt this approach for this post.) Instead of users rating a set of movies on a continuous scale, they simply tell you whether they like a movie or not, and the RBM will try to discover latent factors that can explain the activation of these movie choices.

More technically, a Restricted Boltzmann Machine is a stochastic neural network (neural networkmeaning we have neuron-like units whose binary activations depend on the neighbors they’re connected to; stochastic meaning these activations have a probabilistic element) consisting of:

  • One layer of visible units (users’ movie preferences whose states we know and set);
  • One layer of hidden units (the latent factors we try to learn); and
  • A bias unit (whose state is always on, and is a way of adjusting for the different inherent popularities of each movie).

Furthermore, each visible unit is connected to all the hidden units (this connection is undirected, so each hidden unit is also connected to all the visible units), and the bias unit is connected to all the visible units and all the hidden units. To make learning easier, we restrict the network so that no visible unit is connected to any other visible unit and no hidden unit is connected to any other hidden unit.

For example, suppose we have a set of six movies (Harry Potter, Avatar, LOTR 3, Gladiator, Titanic, and Glitter) and we ask users to tell us which ones they want to watch. If we want to learn two latent units underlying movie preferences – for example, two natural groups in our set of six movies appear to be SF/fantasy (containing Harry Potter, Avatar, and LOTR 3) and Oscar winners (containing LOTR 3, Gladiator, and Titanic), so we might hope that our latent units will correspond to these categories – then our RBM would look like the following:


(Note the resemblance to a factor analysis graphical model.)

State Activation

Restricted Boltzmann Machines, and neural networks in general, work by updating the states of some neurons given the states of others, so let’s talk about how the states of individual units change. Assuming we know the connection weights in our RBM (we’ll explain how to learn these below), to update the state of unit  i :

  • Compute the activation energy  ai=∑jwijxj  of unit  i , where the sum runs over all units  j that unit  i  is connected to,  wij  is the weight of the connection between  i  and  j , and  xj  is the 0 or 1 state of unit  j . In other words, all of unit  i ’s neighbors send it a message, and we compute the sum of all these messages.
  • Let  pi=σ(ai) , where  σ(x)=1/(1+exp(−x))  is the logistic function. Note that  pi  is close to 1 for large positive activation energies, and  pi  is close to 0 for negative activation energies.
  • We then turn unit  i  on with probability  pi , and turn it off with probability  1−pi .
  • (In layman’s terms, units that are positively connected to each other try to get each other to share the same state (i.e., be both on or off), while units that are negatively connected to each other are enemies that prefer to be in different states.)

For example, let’s suppose our two hidden units really do correspond to SF/fantasy and Oscar winners.

  • If Alice has told us her six binary preferences on our set of movies, we could then ask our RBM which of the hidden units her preferences activate (i.e., ask the RBM to explain her preferences in terms of latent factors). So the six movies send messages to the hidden units, telling them to update themselves. (Note that even if Alice has declared she wants to watch Harry Potter, Avatar, and LOTR 3, this doesn’t guarantee that the SF/fantasy hidden unit will turn on, but only that it will turn on with high probability. This makes a bit of sense: in the real world, Alice wanting to watch all three of those movies makes us highly suspect she likes SF/fantasy in general, but there’s a small chance she wants to watch them for other reasons. Thus, the RBM allows us to generate models of people in the messy, real world.)
  • Conversely, if we know that one person likes SF/fantasy (so that the SF/fantasy unit is on), we can then ask the RBM which of the movie units that hidden unit turns on (i.e., ask the RBM to generate a set of movie recommendations). So the hidden units send messages to the movie units, telling them to update their states. (Again, note that the SF/fantasy unit being on doesn’t guarantee that we’ll always recommend all three of Harry Potter, Avatar, and LOTR 3 because, hey, not everyone who likes science fiction liked Avatar.)

Learning Weights

So how do we learn the connection weights in our network? Suppose we have a bunch of training examples, where each training example is a binary vector with six elements corresponding to a user’s movie preferences. Then for each epoch, do the following:

  • Take a training example (a set of six movie preferences). Set the states of the visible units to these preferences.
  • Next, update the states of the hidden units using the logistic activation rule described above: for the  j th hidden unit, compute its activation energy  aj=∑iwijxi , and set  xj  to 1 with probability  σ(aj)  and to 0 with probability  1−σ(aj) . Then for each edge  eij , compute  Positive(eij)=xi∗xj  (i.e., for each pair of units, measure whether they’re both on).
  • Now reconstruct the visible units in a similar manner: for each visible unit, compute its activation energy  ai , and update its state. (Note that this reconstruction may not match the original preferences.) Then update the hidden units again, and compute  Negative(eij)=xi∗xj  for each edge.
  • Update the weight of each edge  eij  by setting  wij=wij+L∗(Positive(eij)−Negative(eij)) , where  L  is a learning rate.
  • Repeat over all training examples.

Continue until the network converges (i.e., the error between the training examples and their reconstructions falls below some threshold) or we reach some maximum number of epochs.

Why does this update rule make sense? Note that

  • In the first phase,  Positive(eij)  measures the association between the  i th and  j th unit that we want the network to learn from our training examples;
  • In the “reconstruction” phase, where the RBM generates the states of visible units based on its hypotheses about the hidden units alone,  Negative(eij)  measures the association that the network itself generates (or “daydreams” about) when no units are fixed to training data.

So by adding  Positive(eij)−Negative(eij)  to each edge weight, we’re helping the network’s daydreams better match the reality of our training examples.

(You may hear this update rule called contrastive divergence, which is basically a funky term for “approximate gradient descent”.)

Examples

I wrote a simple RBM implementation in Python (the code is heavily commented, so take a look if you’re still a little fuzzy on how everything works), so let’s use it to walk through some examples.

First, I trained the RBM using some fake data.

  • Alice: (Harry Potter = 1, Avatar = 1, LOTR 3 = 1, Gladiator = 0, Titanic = 0, Glitter = 0). Big SF/fantasy fan.
  • Bob: (Harry Potter = 1, Avatar = 0, LOTR 3 = 1, Gladiator = 0, Titanic = 0, Glitter = 0). SF/fantasy fan, but doesn’t like Avatar.
  • Carol: (Harry Potter = 1, Avatar = 1, LOTR 3 = 1, Gladiator = 0, Titanic = 0, Glitter = 0). Big SF/fantasy fan.
  • David: (Harry Potter = 0, Avatar = 0, LOTR 3 = 1, Gladiator = 1, Titanic = 1, Glitter = 0). Big Oscar winners fan.
  • Eric: (Harry Potter = 0, Avatar = 0, LOTR 3 = 1, Gladiator = 1, Titanic = 1, Glitter = 0). Oscar winners fan, except for Titanic.
  • Fred: (Harry Potter = 0, Avatar = 0, LOTR 3 = 1, Gladiator = 1, Titanic = 1, Glitter = 0). Big Oscar winners fan.

The network learned the following weights:

                 Bias Unit       Hidden 1        Hidden 2
Bias Unit       -0.08257658     -0.19041546      1.57007782 
Harry Potter    -0.82602559     -7.08986885      4.96606654 
Avatar          -1.84023877     -5.18354129      2.27197472 
LOTR 3           3.92321075      2.51720193      4.11061383 
Gladiator        0.10316995      6.74833901     -4.00505343 
Titanic         -0.97646029      3.25474524     -5.59606865 
Glitter         -4.44685751     -2.81563804     -2.91540988

Note that the first hidden unit seems to correspond to the Oscar winners, and the second hidden unit seems to correspond to the SF/fantasy movies, just as we were hoping.

What happens if we give the RBM a new user, George, who has (Harry Potter = 0, Avatar = 0, LOTR 3 = 0, Gladiator = 1, Titanic = 1, Glitter = 0) as his preferences? It turns the Oscar winners unit on (but not the SF/fantasy unit), correctly guessing that George probably likes movies that are Oscar winners.

What happens if we activate only the SF/fantasy unit, and run the RBM a bunch of different times? In my trials, it turned on Harry Potter, Avatar, and LOTR 3 three times; it turned on Avatar and LOTR 3, but not Harry Potter, once; and it turned on Harry Potter and LOTR 3, but not Avatar, twice. Note that, based on our training examples, these generated preferences do indeed match what we might expect real SF/fantasy fans want to watch.

Modifications

I tried to keep the connection-learning algorithm I described above pretty simple, so here are some modifications that often appear in practice:

  • Above,  Negative(eij)  was determined by taking the product of the  i th and  j th units after reconstructing the visible units once and then updating the hidden units again. We could also take the product after some larger number of reconstructions (i.e., repeat updating the visible units, then the hidden units, then the visible units again, and so on); this is slower, but describes the network’s daydreams more accurately.
  • Instead of using  Positive(eij)=xi∗xj , where  xi  and  xj  are binary 0 or 1 states, we could also let  xi  and/or  xj  be activation probabilities. Similarly for  Negative(eij) .
  • We could penalize larger edge weights, in order to get a sparser or more regularized model.
  • When updating edge weights, we could use a momentum factor: we would add to each edge a weighted sum of the current step as described above (i.e.,  L∗(Positive(eij)−Negative(eij) ) and the step previously taken.
  • Instead of using only one training example in each epoch, we could use batches of examples in each epoch, and only update the network’s weights after passing through all the examples in the batch. This can speed up the learning by taking advantage of fast matrix-multiplication algorithms.

Further

If you’re interested in learning more about Restricted Boltzmann Machines, here are some good links.

 Jul 18th, 2011  expository

受限玻尔兹曼机的英文详细介绍(论文) 详细介绍了受限玻尔兹曼机的原理,需要有一定的英文和数学功底。 立即下载

相关推荐

A Practical Guide to Training Restricted Boltzmann Machines

A Practical Guide to Training Restricted Boltzmann Machines一文详细诠释了rbm的训练过程,有深度学习专家Hiton撰写,对于深度学习的学习很有帮助!

受限波兹曼机导论Introduction to Restricted Boltzmann Machines

Suppose you ask a bunch of users to rate a set of movies on a 0-100 scale. In classicalfactor analysis, you could then try to explain each movie and user in terms of a set of latentfactors. For exam...

weixin_30505043的博客 212

受限波尔兹曼机(Restricted Boltzmann Machines)介绍

RBM在深度学习(deep learning)算法中有着非常重要的应用,本文介绍了RBM的基本概念,并介绍了几种有代表性的算法。作者西安交大张春霞,姬楠楠,王冠伟。

受限玻尔兹曼机(RBM,Restricted Boltzmann Machines)浅介

本文是我在阅读[1]之后做的一个读书笔记,所以这里的内容几乎也是翻译外加一些自己的理解,希望对读者有益。     概括地说,RBM根据MLE原理来估计预定义分布中的参数,以便预定义分布能尽可能地逼近产生观测数据的未知分布。多个RBM分层堆叠而成的DBN(deep belief networks)构成深度学习的主要框架。    RBM是一个随机无向图模型(上图),分为可见层和隐层

GrazyThinking的专栏 2162

印尼稻米收成与粮食安全数据集2026年

本数据集收录印度尼西亚2026年水稻收成与粮食安全遥测数据,包含10000条稻米农业收成产量与粮食安全监测记录,围绕印尼国家粮食安全体系中的核心作物水稻展开。稻米是印尼逾两亿人口的主粮,其收成波动直接影响国家粮食安全与进口政策,本数据集将种植区域、收成产量、气候与土壤相关遥测指标整合为结构化表格,可支撑粮食产量预测、收成影响因素分析、区域粮食安全评估与农业政策模拟等研究任务。典型应用场景包括:农业时间序列产量预测模型构建、气候因子与水稻收成的关联分析、省份与地区层面的粮食自给率测算、粮食安全风险预警指标设计,也可作为农业经济、粮食政策与数据科学交叉课程的教学素材。数据为2026年最新版,规模约1.2MB、单CSV结构规整,字段命名清晰,适合Python、Pandas、Scikit-learn等主流工具链直接使用,便于开展探索性分析、可视化与机器学习全流程练习。重要说明:本数据为遥测整理数据,分析结论以数据集文档口径为准,农业决策请结合官方统计综合研判。许可:CC0公有领域,可自由使用与二次分发。

企业数字化转型AI大模型赋能智能体平台及AI场景应用规划方案(生态合作版).pptx

企业数字化转型AI大模型赋能智能体平台及AI场景应用规划方案(生态合作版).pptx

深度学习1-1——小白初识RBM

深度学习1-1–小白初识RBM仅以此系列记录&分享经典论文《An Introduction to Restricted Boltzmann Machines》从BM(玻尔兹曼机)说起“玻尔兹曼”分布,其实是一个化学上的概念——“热平衡在物理学领域通常指温度在时间或空间上的稳定。”在统计学习中,如果我们将需要学习的模型看成高温物体,将学习的过程看成一个降温达到热平衡的过程,最终模型的能量将会收敛为一个

qq_21842193的博客 621

智慧环保生态保护行业可信数据空间应用规划方案(技术方案版).pptx

智慧环保生态保护行业可信数据空间应用规划方案(技术方案版).pptx

限制玻尔兹曼机导论Introduction to Restricted Boltzmann Machines

Introduction to Restricted Boltzmann Machines Suppose you ask a bunch of users to rate a set of movies on a 0-100 scale. In classical factor analysis, you could then try to explain each m

GarfieldEr007的专栏 1432

气象服务行业可信数据空间应用规划方案(技术方案版).pptx

气象服务行业可信数据空间应用规划方案(技术方案版).pptx

机器视觉检测实战教程 19页成品课程光源选型Halcon标定海康相机VisionPro

内容概要:机器视觉检测实战教程,19页6章成品课程PDF:视觉检测光源选型对照表+五个翻车案例、Halcon模板匹配create_shape_model七参数整定、视觉标定三类(9点/旋转/手眼)、海康工业相机C#取图黑图丢帧排坑、VisionPro QuickBuild部署License迁移C#宿主、远心镜头选型。含C#/Halcon代码与现场案例。适用人群:机器视觉工程师、自动化检测开发、企业内训。使用场景:视觉方案设计、选型、标定部署、团队培训。原创整理。

Learning Latent Dynamics for Planning from Pixels 从像素中学习用于规划的潜在动态

Abstract摘要规划在已知环境动力学的控制任务中取得了巨大成功。为了在未知环境中利用规划,智能体需要从与世界的交互中学习动力学。然而,学习足够准确以用于规划的动力学模型一直是一项长期挑战,尤其是在基于图像的领域中。我们提出了 Deep Planning Network(PlaNet),这是一种纯基于模型的智能体,能够从图像中学习环境动力学,并通过潜在空间中的快速在线规划选择动作。为实现高性能,动力学模型必须能够准确预测未来多个时间步的奖励。

u013250861的博客 265

【医疗器械制造】基于酸碱腐蚀环境的一次性乳胶手套铝合金手模失效机理分析与量产风险控制

内容概要:本文围绕一次性医用乳胶手套生产中使用的铝合金手模在酸碱环境下的失效风险展开系统性分析,重点探讨其在外表面腐蚀、清洗循环、氯介质侵入及表面粗糙度控制等方面的工程挑战。研究表明,尽管市场上已有带涂层的铝合金手模供应,但其能否在强碱、酸洗、毛刷摩擦、冷热交变和氯共存的复杂工况下长期维持Ra 1.0—2.5 μm的表面质量并实现稳定脱模,仍需经过严格的多阶段验证。常规阳极氧化+PTFE涂层方案已被实测排除,而316L不锈钢筒体在湿氯环境中出现的三个月内腐蚀失效案例提示必须重视内腔隔离设计。文章强调,采购决策应基于“小批量试制—离线加速试验—影子生产验证—条件性放行”的四阶段准入机制,结合表面参数、工艺相容性、手套成品性能与全生命周期成本进行综合评估,避免依赖供应商宣传数据。; 适合人群:从事医疗器械制造、材料工程、设备采购与工艺质量管理的技术人员及决策管理者,尤其是涉及乳胶手套生产线优化与手模选型的相关从业人员。; 使用场景及目标:①评估铝合金手模替代陶瓷手模的技术可行性与风险边界;②建立手模选型的系统验证流程,防范因表面失效导致的脱模异常、手套缺陷与非计划停机;③指导采购合同中技术条款的制定,确保供应商交付具备可验证数据支撑; 阅读建议:此资源以工程实证为导向,建议读者结合自身产线工艺参数,重点关注外表面Ra控制、内腔双密封设计、影子生产验证方法及失效链追溯体系的构建,在试制阶段严格执行多维度联合判定标准,审慎推进规模化替代。

【DBO三维路径规划】基于多策略改进的蜣螂算法MSDBO多无人机协同集群避障路径规划(目标函数:最低成本:路径、高度、威胁、转角)研究(Matlab代码实现)

内容概要:本文研究了基于多策略改进的蜣螂优化算法(MSDBO)在多无人机协同集群三维路径规划中的应用,旨在解决复杂三维环境下的避障与最优路径搜索问题。通过构建包含路径长度、飞行高度、威胁规避和转弯角度的综合成本目标函数,利用MSDBO算法对多无人机系统的飞行轨迹进行全局优化,实现在保证安全性与稳定性的前提下达到最低能耗与最优性能的目标。文中详细阐述了算法的多策略改进机制、三维环境建模方法以及多无人机协同避障策略,并通过Matlab仿真平台验证了该方法相较于传统优化算法在收敛速度、路径质量和鲁棒性方面的优越性能。; 适合人群:具备一定优化算法理论基础和无人机控制系统知识的科研人员及研究生,特别适用于从事智能优化算法、多智能体协同控制、路径规划与自主导航等相关领域的研究人员和技术开发者。; 使用场景及目标:①应用于灾害救援、军事侦察、智能巡检等复杂三维场景中多无人机协同任务的路径规划;②为多目标约束下的路径优化、动态避障与能耗平衡问题提供高效的算法支持与仿真验证工具;③推动智能优化算法在无人系统自主决策与协同控制中的实际应用与工程落地。; 阅读建议:建议读者结合提供的Matlab代码深入理解算法的具体实现过程,重点关注目标函数设计、多策略改进机制(如种群初始化、自适应参数调整、局部搜索增强等)及仿真结果分析部分,同时可尝试将该算法迁移至其他智能优化算法框架中进行对比研究,以深化对路径规划问题本质的理解与实践能力的提升。

MCGS的Http协议驱动,支持http协议

硬件配置 参数设置 设备属性 设备命令 通讯状态 附录 #HTTP 1、 设备简介 驱动类型 该驱动为独立设备,直接添加构件即可,不需要挂载到父设备下 通讯协议 采用标准HTTP1.1协议 通讯方式 驱动构件为客户端。 2、 硬件连接 本软件在通讯之前,请保证通讯连接正确。通讯连接方式:采用标准的RJ45接口的以太网接线(对调网线或使用HUB交换)。 3、 设备通讯参数 本设备为独立设备,无父设备参数 4、 设备构件参数设置 “HTTP设备”参数设置如下: <1> 内部属性: 构建url信息。 <2> 最小采集周期:本软件对设备进行操作的时间周期,单位为ms,默认为100ms。 <3> 通讯等待时间:通讯数据接收等待时间,默认设置为200ms,当采集速度要求较高或数据量较大时,设置值可适当减小或增大。 <4> HttpHead:http head添加需要的字段,不需要添加http head时值为空,默认不添加。 5、 设备命令 本设备构件提供设备命令,用于进行相应的操作,设备命令的格式如下: 设备命令 命令介绍 GetHttpHead 格式1 GetHttpHead(headStr) 用途 获取http head参数 参数 headStr 参数可以为字符型变量、字符串常量 例1 !SetDevice(设备0,6,"GetHttpHead(headStr)") 获取http head数据保存到headStr中 SetHttpHead 格式1 SetHttpHead(headStr) 用途 设置http head参数 参数 headStr 参数可以为字符型变量、字符串常量

NPU、CPU和GPU

NPU、CPU和GPU是三种不同类型的处理器,它们各自有不同的设计特点和应用场景。NPU(Neural Processing Unit,神经网络处理单元)是一种专门为加速神经网络计算而设计的处理器。

m0_60697583的博客 1762
上一篇: linux 使用 c.vim
下一篇: C/C++如何传递二维数组?
overstack
博客等级 码龄14年 115粉丝 55原创
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付:10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值