View source for 蒋玉骅

{{Person
|first=玉骅
|last=蒋
|wp-link=
|birthday=2002-03-28
|died=
|gender=Male
|description=TEEP 0, Xingjian College, Tsinghua University
}}

<noinclude>
[[蒋玉骅 认知基础2021年秋 主页]]

=Basic Info=
{|class=wikitable
|style="width:150px"|'''Name'''||style="width:350px"|蒋玉骅||style="width:150px"|'''Class'''||style="width :350px"| TEEP 0
|-
|'''Date of Birth'''||2002-03-28||'''Gender'''||Male
|-
|'''Institution'''||TEEP 0, Xingjian College, Tsinghua Univ.||'''E-mail'''||jiangyh20@mails.tsinghua.edu.cn
|-
|}

=课程感悟=
第一周：
  key-value pair 揭示了复杂网络系统工作原理的本质。用这样的方式去理解范式转换，就会发现科学革命实际就是key-value pair 的扩增与删减；不可通约性来自命名空间的不同。而这样的不可通约正是历代科学家致力于消除的，这就像科学家们企图寻找科学语言的“罗塞塔石碑”一样。以人工智能为例：微软近期就提出了深度学习框架的通用语言——repo1.0，他们希望通过构建这一深度学习框架“罗塞塔石碑”，让研究者们能够在不同框架之间轻松运用专业知识，并且实现不同开源社区之间的合作。<ref>https://link.zhihu.com/?target=https%3A//github.com/ilkarman/DeepLearningFrameworks</ref>如果这能实现，就可以创建一个跨语言对比的常用设置（Python、Julia、R），实现不同开源社区之间的合作。
第二周：
  课上的视频让我想到Paul Nurse的一句话：生命的深层结构指的是细胞或有机体这样的基本单元，它们能够自我繁殖，并允许微小的变异。<ref>Paul Nurse,''What is Life?'',</ref>繁殖与变异共同通过自然选择推动物种的演化，从而形成多样化的生物种群。它们不仅能够在变化的环境中存活，还能够利用新的机会。而那些成功适应环境的单元就能继续繁殖后代。
  类似的机制在不同的尺度上都发挥着作用，构成了众多关键生物过程的基础。胚胎从单细胞发育为成熟有机体的过程中，会经历好几个生长阶段（人类有几十个），每个都与前一个略有不同。最终，受精卵繁衍出各种不同的细胞，包括心脏、肝脏和脑部的细胞。
  这样的原理成为了冯·诺依曼一些设计的出发点。他曾精确地设计了一个被称为“通用复制器”的数学模型。它包括三个基本组成部分 ：机器A是一台可以根据指令整合资源并进行组装的机器 ；程序B能够指挥机器A ；主程序C可以指挥A来制造A+B+C。<ref>John von Neumann, ''Theory of Self-Reproducing Automata'', </ref>从技术上讲，它是一台元胞自动机，可以从周围随机散落的碎片中获取零件。原则上，根据他的设计，我们可以用现代技术造出一个3D打印与计算机的混合系统，它能够收集材料来制作你想要的东西或复制其自身。通过精心设计故意犯错的程序或宽松的质量控制，我们也能解锁生命的另一个秘密——变异。

==References==
<references />

=科学革命的结构的读后感=
Chapter 2：
  Paradigms are theories created by one of the pre-paradigm schools. Though the author gives description rather than explanation of the arduous process of acquiring paradigms, we can conclude that when some theories are universally accepted, and can attract an enduring group of adherents away from competing modes of scientific activity, paradigms that prove able to guide the whole group’s research emerge all of a sudden.
  The emergence of a paradigm contributes to scientific inquiry in seven aspects. First, it makes both fact collection and theory articulation highly directed activities, so scientists no longer explore nature casually or at random. Second, it suggests which experiments will be worth performing so scientists are confident what they are studying is highly relevant. Third, the end of interschool debate ends the constant reiteration of fundamentals. The confidence that they are on the right track encouraged scientists to undertake more precise, esoteric, and consuming sorts of work. Fourth, it transforms a group previously interested merely in the study of nature into a profession or a discipline. Fifth, it marks the beginning of specific classification. Sixth, it creates advanced systems that improve effectiveness and efficiency, in particular, for esoteric work. Above all, it produces scientific community whose members push on to more concrete and recondite problems, and increasingly they report their results in articles addressed to other electricians.
  Ultimately, the emergence of a paradigm is invariably followed by a truth boom, as truth emerges more readily from error than from confusion.
Chapter 3：
  Although paradigms have shown to be particularly revealing of the nature, the match between facts and theories is still imperfect, and scientists have to avoid approximations and obtain satisfactory agreements.
  To do this, they need to answer the following questions. First, how to conduct empirical works to articulate the paradigm? Second, what further explorations relative to theoretical works can they make to classify theoretical problems of normal science? Third, how to use existing theories to predict factual information of intrinsic value?
  All their efforts serve for the reformulation of a paradigm: to articulate the paradigm elegantly and logically in mathematics, which is both theoretical and experimental. The desire for acknowledgement and the ambition for fame ensure scientists to pursue the same goal. Driven by anthropic ultimate curiosity about the nature, they are committed to solve the problems after the acquisition of the paradigm. 
  The problems always exist, as a successful paradigm is not, however, to be either completely successful with a single problem or notably successful with any large number.
Chapter 4：
   What are characteristics of normal science?
   Normal science shares paralleled characteristics with puzzle-solving. First, it offers challenging problems as puzzles that can test ingenuity or skill in solution and thus drive scientists on. Second, the criterion of these problems is the assured existence of a solution, but on the contrary, has nothing to do with the intrinsic value of their outcome. Third, these puzzles are invariably so attractive that scientists attack them with remarkable passion and devotion, holding the conviction that, if only they are skillful enough, they will succeed in solving a puzzle that no one before has solved or solved so well. It is very much the same thing as a child is immersed in solving crossword puzzles that may not give him any benefit.
   Above all, like puzzle-solving, normal science has rules. In the same way as all the pieces must be used before you solve a jigsaw puzzle, until certain conditions have been satisfied, no problem can be solved. Rules play the role of established viewpoint or preconception that bound the admissible solutions to theoretical problems. Under the influence of rules, scientists adopt the attitude that the results of their researches must fall into a narrow range that the paradigm restricts. As a child must obey rules in his games without fail, all these rules have undoubtedly held for scientists at all times.