淘客熙熙

主题:国债负利率的日本,越借钱政府赚得越多,当然不怕花钱 -- forger

共:💬86 🌺272 🌵9
全看分页树展 · 主题 跟帖
家园 大脑神经网络"量子效應 "

this is an another subject worth of tons of books, again, what we care about is its "core algorism", not really the "technical" details (which is "painfully" a core part of physics as well, so, we all have to be technical evils) of it, for the benefit of coding it out, potentially making a thinking model out of it, that is kind of why I heavily write in physics and math;

1.

【原创】量子化学历史拾零-前言 花23 西行的风 字425 2008-04-26 11:40:23

原创】二 从二战难民到诺贝尔 Walter Kohn 花31 西行的风 字1887 2008-04-28 00:20:13

..O KohnSham 方程突破: 再现原子的壳层结构 花4 晓兵 字9544 2013-08-19 18:06:29

...O 彭桓武: 非周期晶体能级间隔>n倍室温能量kt =遗传稳 晓兵 字551 2013-08-23 17:23:17

...O 计算化学是以后材料学的前景所在 黄土布衣 字131 2013-08-21 23:44:42

....O 科學研究本來就是循證過程 红尘有幸蓝锦夏 字97 2013-08-25 20:58:25

.....O 主要问题是现在材料方面的计算理论还不行 黄土布衣 字54 2013-09-13 22:05:25

......O 就算我歧視吧 開會遇到理論學講演我一律是不聽跑去逛街的 红尘有幸蓝锦夏 字112 2013-09-14 11:23:25

....O trial & error材料研究模式:电子的"理性" 晓兵 字2496 2013-08-22 15:14:02

....O "别人几十年甚至几百年的经验积累都成shit了" 花1 晓兵 字3152 2013-08-22 11:42:41

2.

大脑所有的神经元细胞(1000亿以上), 1 细胞 =1000 Trillion atoms, the building parts of 细胞 such as DNA, RNA, protein, all have millions of atoms each.

so, basically 大脑= a group of atoms working together as a quantum system in terms of physics, with Numbers of atoms > out of current physics model's reach.

if I remember correctly, our current physics model can only handle wave function of a few hundreds of atoms working together as a system.

"本世纪初, 人们创立了量子力学理论, 用于研究微观粒子的运动规律。1927 年海特勒和伦

敦应用量子力学原理研究了两个氢原子组成氢分子的化学键本质, 从而开创了量子化学的先

河。就在量子力学诞生后的1929 年, 当时的剑桥大学数学教授Paul Dirac 声称, 把整个化学归

结成一些数学方程的基本定律已经完全搞清楚了, 唯一的问题是方程太复杂, 以至于不能解这

些方程。Dirac 是赫赫有名的量子力学创始人之一, 他与Schrdinger 共享1933 年物理学诺贝尔

奖。Dirac 所指的方程即是现在人们熟知的研究微观粒子的Dirac 定态方程。由他的这番话足

见求解这些方程的困难。颇有趣的是当时作为剑桥大学数学系研究生的Pople 教授今天却因

为他在发展量子化学计算方法方面的贡献而获得1998 年诺贝尔化学奖。Pople 发展的计算程

序正是用于解决象分子这样复杂的化学体系的薛定谔方程。量子化学的根本问题就是求解分

子体系的薛定谔方程。由此可见, Pople 教授对量子化学发展所做贡献的伟大意义。Pople 教授

早在60 年代就率先把计算机应用于量子化学中, 在半经验量子化学计算方法方面做出了重要

的贡献。这里有大家熟知的用于研究不饱和有机分子的电子光谱的PariseParrPople 方法。

这一被称为PPP 方法的发明人之一就是Pople 教授。但是, 在当时, Pople 教授也象当年的Di

rac 一样, 认为由于对计算能力要求很高, 解决非经验的量子化学计算方法, 即便象求解

Schrdinger 方程的HartreeFock 近似解也是很难达到的。

求解Schrdinger 方程的HartreeFock 方法通常是由两个步骤来完成。第一步是把分子轨

道用一套基函数来展开。根据Boys 的建议, 这些基函数可以选用以原子为中心的高斯型函

数。应该指出, 基函数展开方法使复杂的微分方程变成比较简单的矩阵方程, 而后者特别适

合用计算机进行求解。第二步, 也是最重要的一步, 是用积分方法计算体系的各部分能量, 包

括动能、电子与电子之间的排斥能以及电子之间的交换相关能等部分。对分子体系计算这些

积分的工作量随体系电子数N 的增加而以N

4 剧增, 使得对于含多个原子的比较复杂分子的

计算无法实现"

3.

what is the point?

the point is our current chemistry/biology theory about "含多个原子的比较复杂分子" working together as "gene", "protein" system is still largely an experience based science with potentially many fallacies evils fooling around, whom we have no idea, kind of the way chairman X may not like china "mathematics" VP Li, for the same reason, nobody likes "unknown";

the physics layer of GMO is "unknown", period.

and because we don't have a 量子化学计算 model of gene, protein, or any GMO stuff etc=GMO has a huge volatility risk, either directions, or any directions;

but, who cares about that "physics" risk of GMO now? it is already in our mind and body, for good or bad, just swallow it, and good luck (:).

4.

what is the heck of that 量子化学 stuff in relation to 大脑神经网络"量子效應 "? how can I make my brain get rid of the damned "markov chain" thing, so I can think outside of box and be smarter?

well, "量子效應 " is much more related to 自旋 of 電子, atoms, etc, than anything else, with dirac equation breaking the ice for his fellow human being in getting further into "量子效應 " world, and 自旋 is both local and global, as such we have gauge field theory and model, pioneered by prof 杨振宁;

论AB效应解释的语境依赖性- 豆丁网

www.docin.com 文学/艺术/军事/历史 文物考古

轉為繁體網頁

2012年3月26日 - ... 方式,是由吴大骏和杨振宁于1975年所提出的,既不使用电磁场术语,也 ... 与场强和矢势的性质所不同,狄拉克相因子s(c)不是在特定的时空中 ...

5.

connections of human being is one of the tricks to trigger possibly 大脑神经网络"量子效應 "

孙昌璞院士做客大家讲坛谈量子效应与器件_学院新闻_山东大学物理 ...

www.phym.sdu.edu.cn/2011/1031/1413.html

轉為繁體網頁

2011年10月31日 - 报告中孙昌璞院士首先介绍了自己在光合作用中量子效应. ... 另外,孙院士还介绍了动物大规模的迁徙问题。 ... 这是一个涉及到量子力学一些基本问题的过程,即宏观微弱量对量子演化过程的影响会不会使初态 ... 孙昌璞,中国科学院院士。1996年起,任中科院理论物理所研究员。2009年11月当选为中国科学院院士。

孙昌璞 is an another student of prof 杨振宁, who taught 孙昌璞 群表示论 "trick", etc

Born-Oppenheimer近似的精确验证与经典对应张力,王成,孙昌璞(东北师范大学理论 ... 关于q一Bose代数B,人们虽然已构造了许多具体的表示,但对于表示论的基本 ...

and as I posted before, 杨振宁's father gave 杨振宁 a critical short but well written 群论 book when 杨振宁 was working on a paper assigned by 吴大骏 on some kind of "atom spectrum" problems , a book benefited him for his life, as 杨振宁 said;

6.

now with the emerging AI economy, you put 杨振宁's father, 孙昌璞 etc all in a global 3D grand classroom virtually all over the globe, teaching all the young brains of the whole world, making n% of them all become 中国科学院院士 level scientists, where this world will be in terms of whatever progresses?

7.

"量子效應 "/自旋 as an non-linear "breaking through" quantum effect of both locally & globally through of connected human 大脑神经网络 is actually the trick of 列宁建党建军组织 model, and with that model, TG powered numerous 生的伟大 死的光荣 folks, it is essentially a 自旋/"量子效應 ", or 精神原子弹;

goog, fb, amzn, tsla=kind of uncle sam's 精神原子弹 or 自旋/"量子效應 ", in a way, in today's AI war;

and in that AI platform, we can possible have our brain get rid of the damned "馬可夫鏈(Markov Chain) " and become much more productive, although we may have to pay goog etc some kind of fees.

again, I would hope that chairman X and his red gen II club buddies are not going to fight a yesterday's war of "生的伟大 死的光荣" in today's world, that would be a losing trade with potentially permanent damage to Chinese as a nation, period.

通宝推:墨虎,
全看分页树展 · 主题 跟帖


有趣有益,互惠互利;开阔视野,博采众长。
虚拟的网络,真实的人。天南地北客,相逢皆朋友

Copyright © cchere 西西河