当前位置:网站首页>熵与形态的非递进现象
熵与形态的非递进现象
2022-07-27 10:26:00 【黑榆】

现在假如有两个盒子,这两个盒子里有两种不同的粒子A和B,现在假设A排列成0的形状,B排列成1的形状。把两个盒子作为一个整体计算这个整体的熵。让这个熵的为H01,现在让B粒子由1的形状变成2的形状,此时的熵为H02,那H01和H02哪个更大?

B分子由于随机运动恰好由1的形态变成了2的形态,这个过程是熵增的还是熵减的?
现在假设A粒子不是0的形状,而是3的形状。再让B粒子由1随机运动成2,此时两个盒子的熵的变化量为H32-H31

则此时H02-H01与H32-H31之间是什么关系?难道他们相等么?
他们应该是不相等的,如果认为H02-H01 = H32-H31,就是认为这个变化的过程只有B粒子运动,则整个过程的熵仅与B粒子有关,A存在与否或者无论A的形态是什么对整个过程没有任何关系。
但这与前提矛盾,前提是要把A和B看作一个整体,并计算这个整体的熵的变化。因此B粒子形态的变化应该是相对A粒子分布形态的变化,有理由认为
(3,1)---m*n*k---(1,0)(0,1)
(3,2)---m*n*k---(1,0)(0,1)
用神经网络分类3和1,3和2得到的迭代次数n31,n32与这两个状态的熵成反比

按照前面得到的形态数轴
8 | 6 | 1 | 9 | 3 | 4 | 2 | 7 | 5 | |
0 | 27905.07 | 25605.5 | 23558.45 | 13346.79 | 12572.23 | 11983.15 | 8358.603 | 7822.01 | 5402.955 |
1 | 2 | 9 | 7 | 8 | 4 | 0 | 5 | 6 | |
3 | 35671.24 | 33389.61 | 19919.65 | 17331.39 | 17015.68 | 15199.52 | 12572.23 | 11703.08 | 8136.266 |
常数c应该与网络结构有关,设为1

计算表明这两个过程熵的变化量不同,且这两个过程很巧合的都是熵增的过程,并且B粒子相对于0运动的熵增量更大。
8 | 6 | 1 | 9 | 3 | 4 | 2 | 7 | 5 | |
0 | 27905.07 | 25605.5 | 23558.45 | 13346.79 | 12572.23 | 11983.15 | 8358.603 | 7822.01 | 5402.955 |
3 | 0 | 8 | 2 | 9 | 6 | 5 | 4 | 7 | |
1 | 35671.24 | 23558.45 | 16861 | 11792.54 | 10721.14 | 10241.39 | 10137.68 | 9577.513 | 9568.94 |
3 | 4 | 9 | 7 | 1 | 6 | 0 | 8 | 5 | |
2 | 33389.61 | 19984.86 | 13772.15 | 12555.62 | 11792.54 | 9360.106 | 8358.603 | 7658.015 | 7100.643 |
1 | 2 | 9 | 7 | 8 | 4 | 0 | 5 | 6 | |
3 | 35671.24 | 33389.61 | 19919.65 | 17331.39 | 17015.68 | 15199.52 | 12572.23 | 11703.08 | 8136.266 |
2 | 9 | 3 | 0 | 1 | 6 | 7 | 8 | 5 | |
4 | 19984.86 | 18523.66 | 15199.52 | 11983.15 | 9577.513 | 9020.96 | 7572.704 | 6106.347 | 5689.266 |
3 | 1 | 7 | 2 | 9 | 8 | 4 | 0 | 6 | |
5 | 11703.08 | 10137.68 | 8617.161 | 7100.643 | 6794.688 | 6116.397 | 5689.266 | 5402.955 | 5362.608 |
0 | 1 | 2 | 9 | 4 | 7 | 8 | 3 | 5 | |
6 | 25605.5 | 10241.39 | 9360.106 | 9044.211 | 9020.96 | 8983.447 | 8626.678 | 8136.266 | 5362.608 |
9 | 3 | 2 | 1 | 6 | 5 | 0 | 4 | 8 | |
7 | 20211.46 | 17331.39 | 12555.62 | 9568.94 | 8983.447 | 8617.161 | 7822.01 | 7572.704 | 7073.432 |
0 | 3 | 1 | 6 | 2 | 7 | 9 | 5 | 4 | |
8 | 27905.07 | 17015.68 | 16861 | 8626.678 | 7658.015 | 7073.432 | 6966.322 | 6116.397 | 6106.347 |
7 | 3 | 4 | 2 | 0 | 1 | 6 | 8 | 5 | |
9 | 20211.46 | 19919.65 | 18523.66 | 13772.15 | 13346.79 | 10721.14 | 9044.211 | 6966.322 | 6794.688 |
但这或许仅仅是个巧合因为如果A的形态是4,7,9,B粒子由1变化为2这个过程是熵减的,所以粒子由1随机运动为2如果相对于0是有序变无序,而如果相对于9是无序变有序。
边栏推荐
- ECCV 2022 | 同时完成四项跟踪任务!Unicorn: 迈向目标跟踪的大统一
- Webrtc realizes simple audio and video call function
- [intensive reading of thesis]bert
- 华硕无双,这可能是屏幕最好的平价高刷轻薄笔记本
- flask_restful中的输出域(Resource、fields、marshal、marshal_with)
- It is thought-provoking: is syntax really important? Qiu Xipeng group proposed a powerful baseline for aspect based emotional analysis
- 分布式块设备复制:客户端
- Your appearance is amazing! Two JSON visualization tools are recommended for use with swagger. It's really fragrant
- How to smooth the online and offline of Web Services
- File upload vulnerability bypass method
猜你喜欢

YonBuilder赋能创新,用友第四届开发者大赛“金键盘奖”开启竞逐!

ECCV 2022 | 同时完成四项跟踪任务!Unicorn: 迈向目标跟踪的大统一

Multipoint bidirectional republication and routing strategy

Analysis of heterogeneous computing technology

Overview of user space lock on mobile platform

The largest square of leetcode (violence solving and dynamic programming solving)

Metaaploit post penetration technology knowledge

No identifier specified for entity solution

No Identifier specified for entity的解决办法

北京公示儿童鞋抽查 8组批产品不合格琪尔特登榜
随机推荐
WEB服务如何平滑的上下线
Edata base, a secondary development project based on spark packaging, is introduced
Set up Samba service
Tcp/ip protocol
Metaaploit post penetration technology knowledge
TensorFlow笔记——基本函数及概念
Beijing publicized the spot check of 8 batches of children's shoes, and qierte was listed as unqualified
img src为空或者src不存在,图片出现白色边框
Apache22 and opencms9.5 integrated configuration
Apache cannot start in phpstudy
flask_ Output fields in restful (resources, fields, marshal, marshal_with)
NodeJS中Error: getaddrinfo ENOTFOUND localhost
Chengying, kangaroo cloud one-stop fully automated operation and maintenance steward, is officially open source
A brief introduction to R language pipeline symbols (% >%) and placeholders (.)
Gamer questions
ASP. Net core dependency injection journey: 1. Theoretical concepts
华硕无双,这可能是屏幕最好的平价高刷轻薄笔记本
招聘顶尖人才!旷视科技“MegEagle创视者计划”正式启动
服务器访问速度
游戏玩家问题