Computing a theory of everything | Stephen Wolfram

603,687 views ・ 2010-04-27

TED


请双击下面的英文字幕来播放视频。

翻译人员: Hao Li 校对人员: Vivian Lee
00:16
So I want to talk today about an idea. It's a big idea.
0
16260
3000
接下来,我今天想谈的是一个宏观理念。
00:19
Actually, I think it'll eventually
1
19260
2000
其实,我认为这个构想最终
00:21
be seen as probably the single biggest idea
2
21260
2000
会被视为上个世纪出现过的
00:23
that's emerged in the past century.
3
23260
2000
最伟大的理念
00:25
It's the idea of computation.
4
25260
2000
那就是计算的理念
00:27
Now, of course, that idea has brought us
5
27260
2000
现在,当然,这个理念已经带给我们
00:29
all of the computer technology we have today and so on.
6
29260
3000
所有今天所拥有的电脑科技
00:32
But there's actually a lot more to computation than that.
7
32260
3000
然而,除此之外,还有更多可以计算的事物。
00:35
It's really a very deep, very powerful, very fundamental idea,
8
35260
3000
这真是个非常深刻,非常有用,非常基本的理念
00:38
whose effects we've only just begun to see.
9
38260
3000
而我们只是刚开始见证这个理念的作用
00:41
Well, I myself have spent the past 30 years of my life
10
41260
3000
过去30年里,我致力于
00:44
working on three large projects
11
44260
2000
研究3个大型的项目
00:46
that really try to take the idea of computation seriously.
12
46260
3000
这些项目认真地将计算的理念付诸实践
00:50
So I started off at a young age as a physicist
13
50260
3000
刚开始时我只是个年轻的物理学家
00:53
using computers as tools.
14
53260
2000
运用电脑作为工具
00:55
Then, I started drilling down,
15
55260
2000
然后,我开始深入
00:57
thinking about the computations I might want to do,
16
57260
2000
思考我可能想做的计算
00:59
trying to figure out what primitives they could be built up from
17
59260
3000
尝试找出可以加以演变的主数据类型
01:02
and how they could be automated as much as possible.
18
62260
3000
以及它们尽可能自动运行的方式
01:05
Eventually, I created a whole structure
19
65260
2000
最终,我创立了整个架构
01:07
based on symbolic programming and so on
20
67260
2000
基于符号编程等等
01:09
that let me build Mathematica.
21
69260
2000
然后创造出了Mathematica
01:11
And for the past 23 years, at an increasing rate,
22
71260
2000
过去23年间,以逐年增长的态势
01:13
we've been pouring more and more ideas
23
73260
2000
我们已经为Mathematica注入了
01:15
and capabilities and so on into Mathematica,
24
75260
2000
越来越多的概念和性能
01:17
and I'm happy to say that that's led to many good things
25
77260
3000
而且我很高兴地说这带来了很多进步
01:20
in R & D and education,
26
80260
2000
在研发和教育
01:22
lots of other areas.
27
82260
2000
以及其他很多方面
01:24
Well, I have to admit, actually,
28
84260
2000
当然,我必须承认,事实上
01:26
that I also had a very selfish reason for building Mathematica:
29
86260
3000
我开发Mathematica也有个自私的原因
01:29
I wanted to use it myself,
30
89260
2000
那就是我想要用它
01:31
a bit like Galileo got to use his telescope
31
91260
2000
就像伽利略在400年前
01:33
400 years ago.
32
93260
2000
想要用望远镜一样
01:35
But I wanted to look not at the astronomical universe,
33
95260
3000
但我想了解的不是天文宇宙
01:38
but at the computational universe.
34
98260
3000
而是可计算空间
01:41
So we normally think of programs as being
35
101260
2000
通常我们觉得程序是
01:43
complicated things that we build
36
103260
2000
复杂的东西
01:45
for very specific purposes.
37
105260
2000
我们编程有很多特定的目的
01:47
But what about the space of all possible programs?
38
107260
3000
然而所有程序的空间又有多少呢?
01:50
Here's a representation of a really simple program.
39
110260
3000
这里有个非常简单的程序
01:53
So, if we run this program,
40
113260
2000
所以呢,如果我们运行这个程序
01:55
this is what we get.
41
115260
2000
这就是我们得到的结果
01:57
Very simple.
42
117260
2000
很简单
01:59
So let's try changing the rule
43
119260
2000
接下来,我们稍微修改一下
02:01
for this program a little bit.
44
121260
2000
这个程序的规则
02:03
Now we get another result,
45
123260
2000
我们便得到了另一个结果
02:05
still very simple.
46
125260
2000
仍旧非常简单
02:07
Try changing it again.
47
127260
3000
再试着改一下
02:10
You get something a little bit more complicated.
48
130260
2000
你就看到稍微复杂一点的东西
02:12
But if we keep running this for a while,
49
132260
2000
不过如果我们把这个程序继续运行下去
02:14
we find out that although the pattern we get is very intricate,
50
134260
3000
我们将发现,尽管我们获得的图案十分复杂
02:17
it has a very regular structure.
51
137260
3000
但它具有有规律的结构
02:20
So the question is: Can anything else happen?
52
140260
3000
接下来的问题是:还能发生什么?
02:23
Well, we can do a little experiment.
53
143260
2000
好,我们可以做个小实验
02:25
Let's just do a little mathematical experiment, try and find out.
54
145260
3000
来做个小的数学实验,试着找出规律
02:29
Let's just run all possible programs
55
149260
3000
运行我们所关注的特定总类的
02:32
of the particular type that we're looking at.
56
152260
2000
所有可能的程序
02:34
They're called cellular automata.
57
154260
2000
他们被称为单元自动机
02:36
You can see a lot of diversity in the behavior here.
58
156260
2000
你能看到这里有各种各样的图案模式
02:38
Most of them do very simple things,
59
158260
2000
大多数都很简单
02:40
but if you look along all these different pictures,
60
160260
2000
但是,如果你注意所有不同的图片
02:42
at rule number 30,
61
162260
2000
在30号规则上
02:44
you start to see something interesting going on.
62
164260
2000
你开始看见一些有趣的东西出现
02:46
So let's take a closer look
63
166260
2000
所以我们仔细看一下
02:48
at rule number 30 here.
64
168260
2000
在30号规则这里
02:50
So here it is.
65
170260
2000
就在这里
02:52
We're just following this very simple rule at the bottom here,
66
172260
3000
我们只是按照底部非常简单的规律
02:55
but we're getting all this amazing stuff.
67
175260
2000
然而我们得到了惊人的结果
02:57
It's not at all what we're used to,
68
177260
2000
这与我们过去习惯的事物完全不同
02:59
and I must say that, when I first saw this,
69
179260
2000
而且,我必须说,当我第一次看见它的时候
03:01
it came as a huge shock to my intuition.
70
181260
3000
它让我直觉为之震惊
03:04
And, in fact, to understand it,
71
184260
2000
实际上,为了理解它
03:06
I eventually had to create
72
186260
2000
我们最终不得不建立
03:08
a whole new kind of science.
73
188260
2000
一套全新的科学
03:11
(Laughter)
74
191260
2000
(笑声)
03:13
This science is different, more general,
75
193260
3000
这套科学是与众不同的,并且更加广义的
03:16
than the mathematics-based science that we've had
76
196260
2000
比起已经存在的基于数学的其他科学来说
03:18
for the past 300 or so years.
77
198260
3000
在过去300年甚至更久的时间内
03:21
You know, it's always seemed like a big mystery:
78
201260
2000
你知道的,它总是看似神秘
03:23
how nature, seemingly so effortlessly,
79
203260
3000
自然毫不费力地
03:26
manages to produce so much
80
206260
2000
制造出如此多的东西
03:28
that seems to us so complex.
81
208260
3000
让我们觉得如此复杂
03:31
Well, I think we've found its secret:
82
211260
3000
于是,我觉得我们已经发现了其中的奥秘
03:34
It's just sampling what's out there in the computational universe
83
214260
3000
这只是我们能探索的计算空间的一个样本
03:37
and quite often getting things like Rule 30
84
217260
3000
它们都像30号规则
03:40
or like this.
85
220260
3000
或者像这个
03:44
And knowing that starts to explain
86
224260
2000
在知道这件事后,我们可以开始解释
03:46
a lot of long-standing mysteries in science.
87
226260
3000
很多科学中长期以来的谜团
03:49
It also brings up new issues, though,
88
229260
2000
不过,它也带来新的问题
03:51
like computational irreducibility.
89
231260
3000
就像计算的不可化归性
03:54
I mean, we're used to having science let us predict things,
90
234260
3000
我的意思是我们曾习惯让科学帮我们预测一些事情
03:57
but something like this
91
237260
2000
但是像这样的事情
03:59
is fundamentally irreducible.
92
239260
2000
是根本不可简化的
04:01
The only way to find its outcome
93
241260
2000
发现它结果的唯一方法
04:03
is, effectively, just to watch it evolve.
94
243260
3000
实际上就是看着它演化
04:06
It's connected to, what I call,
95
246260
2000
与之相关的便是我所谓的
04:08
the principle of computational equivalence,
96
248260
2000
计算等价性原则
04:10
which tells us that even incredibly simple systems
97
250260
3000
它告诉我们即使超级简单的系统
04:13
can do computations as sophisticated as anything.
98
253260
3000
也能做极端复杂的计算
04:16
It doesn't take lots of technology or biological evolution
99
256260
3000
不需要多先进的技术或是生物进化过程
04:19
to be able to do arbitrary computation;
100
259260
2000
就能使得它能够做任意的计算
04:21
just something that happens, naturally,
101
261260
2000
这就是自然发生的事情
04:23
all over the place.
102
263260
2000
随处可见
04:25
Things with rules as simple as these can do it.
103
265260
3000
有如此简单规则的东西能达此目的
04:29
Well, this has deep implications
104
269260
2000
而且,这件事有深刻的意义
04:31
about the limits of science,
105
271260
2000
涉及科学的极限
04:33
about predictability and controllability
106
273260
2000
概率论和控制论等
04:35
of things like biological processes or economies,
107
275260
3000
在生物进程或者经济方面发挥作用
04:38
about intelligence in the universe,
108
278260
2000
还有关于宇宙中的智能
04:40
about questions like free will
109
280260
2000
关于自由意志
04:42
and about creating technology.
110
282260
3000
以及创新技术的问题
04:45
You know, in working on this science for many years,
111
285260
2000
从事这些科学工作很多年后
04:47
I kept wondering,
112
287260
2000
我开始思考
04:49
"What will be its first killer app?"
113
289260
2000
第一个令人震惊的应用程序是什么?
04:51
Well, ever since I was a kid,
114
291260
2000
恩,甚至我还是孩子时
04:53
I'd been thinking about systematizing knowledge
115
293260
2000
我就想过关于知识系统化的问题
04:55
and somehow making it computable.
116
295260
2000
以及怎么让它变得可计算
04:57
People like Leibniz had wondered about that too
117
297260
2000
莱布尼兹之辈也已经想过这个问题
04:59
300 years earlier.
118
299260
2000
在300年前
05:01
But I'd always assumed that to make progress,
119
301260
2000
但是我总是假设,为了进步,
05:03
I'd essentially have to replicate a whole brain.
120
303260
3000
我不得不克隆出整个大脑
05:06
Well, then I got to thinking:
121
306260
2000
而现在,我想的是
05:08
This scientific paradigm of mine suggests something different --
122
308260
3000
我的科学模式意味着不一样的东西。
05:11
and, by the way, I've now got
123
311260
2000
并且,顺便提一下,我已经
05:13
huge computation capabilities in Mathematica,
124
313260
3000
使Mathematica具备了超强的计算能力
05:16
and I'm a CEO with some worldly resources
125
316260
3000
并且,我是公司的首席执行官,拥有大量的资源
05:19
to do large, seemingly crazy, projects --
126
319260
3000
来做大型的,看似疯狂的项目。
05:22
So I decided to just try to see
127
322260
2000
所以,我决定尝试知道
05:24
how much of the systematic knowledge that's out there in the world
128
324260
3000
在这世界上,有多少系统化的知识
05:27
we could make computable.
129
327260
2000
是我们能够计算的
05:29
So, it's been a big, very complex project,
130
329260
2000
所以,这是个大型、复杂的项目,
05:31
which I was not sure was going to work at all.
131
331260
3000
我不完全确定它是否可行
05:34
But I'm happy to say it's actually going really well.
132
334260
3000
但是我很高兴地说,它现在进行的不错
05:37
And last year we were able
133
337260
2000
就在去年
05:39
to release the first website version
134
339260
2000
我们发布了第一个网络版本的
05:41
of Wolfram Alpha.
135
341260
2000
Wolfram Alpha
05:43
Its purpose is to be a serious knowledge engine
136
343260
3000
目的是提供一个专业的知识搜索引擎
05:46
that computes answers to questions.
137
346260
3000
它为提问计算答案
05:49
So let's give it a try.
138
349260
2000
所以呢,我们来试试看
05:51
Let's start off with something really easy.
139
351260
2000
让我们先试试简单的东西
05:53
Hope for the best.
140
353260
2000
希望没问题
05:55
Very good. Okay.
141
355260
2000
非常好,没问题
05:57
So far so good.
142
357260
2000
到目前为止,不错
05:59
(Laughter)
143
359260
3000
(笑声)
06:02
Let's try something a little bit harder.
144
362260
3000
让我们试试难一点的东西
06:05
Let's do
145
365260
2000
比如
06:07
some mathy thing,
146
367260
3000
我们做点数学
06:10
and with luck it'll work out the answer
147
370260
3000
希望它能幸运的计算出结果
06:13
and try and tell us some interesting things
148
373260
2000
并且试着告诉我们一些
06:15
things about related math.
149
375260
2000
关于数学的有趣的事
06:17
We could ask it something about the real world.
150
377260
3000
我们可以问他一些现实生活的事情
06:20
Let's say -- I don't know --
151
380260
2000
比如,--- 让我想想 -----
06:22
what's the GDP of Spain?
152
382260
3000
西班牙的国民生产总值是多少?
06:25
And it should be able to tell us that.
153
385260
2000
它应该能告诉我们
06:27
Now we could compute something related to this,
154
387260
2000
现在我们能计算和它相关的事
06:29
let's say ... the GDP of Spain
155
389260
2000
比如西班牙的国民生产总值
06:31
divided by, I don't know,
156
391260
2000
除以, 让我想想
06:33
the -- hmmm ...
157
393260
2000
06:35
let's say the revenue of Microsoft.
158
395260
2000
比如微软公司的收入
06:37
(Laughter)
159
397260
2000
(笑声)
06:39
The idea is that we can just type this in,
160
399260
2000
想法就是我们输入一些好奇的问题
06:41
this kind of question in, however we think of it.
161
401260
3000
不论是什么奇怪的问题
06:44
So let's try asking a question,
162
404260
2000
所以,我们提个问题
06:46
like a health related question.
163
406260
2000
比如有关健康的问题
06:48
So let's say we have a lab finding that ...
164
408260
3000
比如,跟据实验室数据
06:51
you know, we have an LDL level of 140
165
411260
2000
你知道的,有低密度脂蛋白浓度值是140的数据
06:53
for a male aged 50.
166
413260
3000
这是针对50多岁的男性
06:56
So let's type that in, and now Wolfram Alpha
167
416260
2000
我们输入这个,然后Wolfram Alpha
06:58
will go and use available public health data
168
418260
2000
就会使用存在的公共健康数据库
07:00
and try and figure out
169
420260
2000
来试着分析出
07:02
what part of the population that corresponds to and so on.
170
422260
3000
这组数据对应哪部分人群等等
07:05
Or let's try asking about, I don't know,
171
425260
3000
或者我们可以问,让我想想
07:08
the International Space Station.
172
428260
2000
国际空间站的问题
07:10
And what's happening here is that
173
430260
2000
结果就是
07:12
Wolfram Alpha is not just looking up something;
174
432260
2000
Wolfram Alpha不仅在查找信息
07:14
it's computing, in real time,
175
434260
3000
它是在实时计算
07:17
where the International Space Station is right now at this moment,
176
437260
3000
国际空间站现在此刻的位置
07:20
how fast it's going, and so on.
177
440260
3000
它运行的速度等等
07:24
So Wolfram Alpha knows about lots and lots of kinds of things.
178
444260
3000
所以呢,Wolfram Alpha知道很多很多不同的事情
07:27
It's got, by now,
179
447260
2000
到现在为止
07:29
pretty good coverage of everything you might find
180
449260
2000
它几乎可以很好的涵盖了你能在
07:31
in a standard reference library.
181
451260
3000
一个标准图书馆中找到的知识
07:34
But the goal is to go much further
182
454260
2000
不过,我们的目标远不止这些
07:36
and, very broadly, to democratize
183
456260
3000
概括地说
07:39
all of this knowledge,
184
459260
3000
是要使所有的知识民主化
07:42
and to try and be an authoritative
185
462260
2000
并且试着提供
07:44
source in all areas.
186
464260
2000
所有领域中的权威资料
07:46
To be able to compute answers to specific questions that people have,
187
466260
3000
使它能够计算人们特定问题的答案
07:49
not by searching what other people
188
469260
2000
不是靠搜索其他人
07:51
may have written down before,
189
471260
2000
之前可能写下的资料
07:53
but by using built in knowledge
190
473260
2000
而是使用内建知识
07:55
to compute fresh new answers to specific questions.
191
475260
3000
来对特定问题计算新的答案
07:58
Now, of course, Wolfram Alpha
192
478260
2000
现在,当然,Wolfram Alpha
08:00
is a monumentally huge, long-term project
193
480260
2000
是一个非常大型、长远的项目
08:02
with lots and lots of challenges.
194
482260
2000
面临着众多挑战
08:04
For a start, one has to curate a zillion
195
484260
3000
开始的时候,我们要收集数以万计的
08:07
different sources of facts and data,
196
487260
3000
不同的事实来源和数据
08:10
and we built quite a pipeline of Mathematica automation
197
490260
3000
而且,我们建立了Mathematica自动化流水线
08:13
and human domain experts for doing this.
198
493260
3000
还有知识领域专家来做这件事
08:16
But that's just the beginning.
199
496260
2000
不过,这只是开始
08:18
Given raw facts or data
200
498260
2000
对于运用一些没有处理的事实和数据
08:20
to actually answer questions,
201
500260
2000
来解答实际问题
08:22
one has to compute:
202
502260
2000
一方面要计算
08:24
one has to implement all those methods and models
203
504260
2000
另一方面要执行所有的方法、模型
08:26
and algorithms and so on
204
506260
2000
以及算法等等
08:28
that science and other areas have built up over the centuries.
205
508260
3000
而科学以及其他领域于此已发展了数个世纪
08:31
Well, even starting from Mathematica,
206
511260
3000
甚至从Mathematica开始
08:34
this is still a huge amount of work.
207
514260
2000
这仍然是一项浩大工程
08:36
So far, there are about 8 million lines
208
516260
2000
至今为止,有8百万行
08:38
of Mathematica code in Wolfram Alpha
209
518260
2000
Mathematica的代码写在Wolfram Alpha里
08:40
built by experts from many, many different fields.
210
520260
3000
这些代码由很多来自不同领域的专家构建
08:43
Well, a crucial idea of Wolfram Alpha
211
523260
3000
Wolfram Alpha中的一个最重要的想法
08:46
is that you can just ask it questions
212
526260
2000
是你可以问它问题
08:48
using ordinary human language,
213
528260
3000
使用普通人类语言
08:51
which means that we've got to be able to take
214
531260
2000
这意味着我们必须能够接受
08:53
all those strange utterances that people type into the input field
215
533260
3000
人们输入所有的奇怪的文字
08:56
and understand them.
216
536260
2000
并理解它们
08:58
And I must say that I thought that step
217
538260
2000
我必须说我曾觉得做到那一步
09:00
might just be plain impossible.
218
540260
3000
相当不可能
09:04
Two big things happened:
219
544260
2000
后来有了两大重要进步
09:06
First, a bunch of new ideas about linguistics
220
546260
3000
首先是语言学上的很多新想法
09:09
that came from studying the computational universe;
221
549260
3000
来自于对计算空间的研究
09:12
and second, the realization that having actual computable knowledge
222
552260
3000
其次,可计算知识的实现
09:15
completely changes how one can
223
555260
2000
完全地改变了如何一个人能够
09:17
set about understanding language.
224
557260
3000
开始理解语言
09:20
And, of course, now
225
560260
2000
当然,现在
09:22
with Wolfram Alpha actually out in the wild,
226
562260
2000
在浩瀚的网络中有了Wolfram Alpha
09:24
we can learn from its actual usage.
227
564260
2000
我们就能学习它的使用方法
09:26
And, in fact, there's been
228
566260
2000
实际上,一直都有
09:28
an interesting coevolution that's been going on
229
568260
2000
一个有趣的共同进化
09:30
between Wolfram Alpha
230
570260
2000
发生在Wolfram Alpha
09:32
and its human users,
231
572260
2000
和用户之间
09:34
and it's really encouraging.
232
574260
2000
并且,这相当鼓舞人心
09:36
Right now, if we look at web queries,
233
576260
2000
现在,对于任意网络搜索
09:38
more than 80 percent of them get handled successfully the first time.
234
578260
3000
超过百分之80的搜索在第一时间就被成功处理。
09:41
And if you look at things like the iPhone app,
235
581260
2000
如果你看看类似iPhone应用程序的东西
09:43
the fraction is considerably larger.
236
583260
2000
那被成功搜索部分就相当大了
09:45
So, I'm pretty pleased with it all.
237
585260
2000
所以我对此很满意
09:47
But, in many ways,
238
587260
2000
但是,从很多角度看
09:49
we're still at the very beginning with Wolfram Alpha.
239
589260
3000
我们仍然处于Wolfram Alpha开发的初级阶段。
09:52
I mean, everything is scaling up very nicely
240
592260
2000
我的意思是,每件事情的规模都在扩大
09:54
and we're getting more confident.
241
594260
2000
我们也变得更有信心
09:56
You can expect to see Wolfram Alpha technology
242
596260
2000
你能期待看到Wolfram Alpha技术
09:58
showing up in more and more places,
243
598260
2000
在越来越多的地方使用
10:00
working both with this kind of public data, like on the website,
244
600260
3000
既能使用公共数据,比如网站
10:03
and with private knowledge
245
603260
2000
又能使用私人数据
10:05
for people and companies and so on.
246
605260
3000
给个人和公司等等提供服务
10:08
You know, I've realized that Wolfram Alpha actually gives one
247
608260
3000
我觉得Wolfram Alpha其实是一个
10:11
a whole new kind of computing
248
611260
2000
全新的计算方法
10:13
that one can call knowledge-based computing,
249
613260
2000
我们可以称之基于知识的计算
10:15
in which one's starting not just from raw computation,
250
615260
3000
这种计算方法,不仅可以使用原始数据
10:18
but from a vast amount of built-in knowledge.
251
618260
3000
还能使用大量的内建知识
10:21
And when one does that, one really changes
252
621260
2000
而且,一个能做这样计算的工具真的能够改变
10:23
the economics of delivering computational things,
253
623260
3000
传递可计算事物的理论
10:26
whether it's on the web or elsewhere.
254
626260
2000
无论在网络上或者是其他地方
10:28
You know, we have a fairly interesting situation right now.
255
628260
3000
我们现在处于一个很有意思的状态
10:31
On the one hand, we have Mathematica,
256
631260
2000
一方面,我们拥有Mathematica这个软件
10:33
with its sort of precise, formal language
257
633260
3000
它有精确性,正规性
10:36
and a huge network
258
636260
2000
以及大规模
10:38
of carefully designed capabilities
259
638260
2000
设计仔细的功能网络
10:40
able to get a lot done in just a few lines.
260
640260
3000
用几行代码就能做很多事情
10:43
Let me show you a couple of examples here.
261
643260
3000
我来展示几个例子
10:47
So here's a trivial piece of Mathematica programming.
262
647260
3000
这是Mathematica编程中很小的一段代码
10:51
Here's something where we're sort of
263
651260
2000
这里是我们整合
10:53
integrating a bunch of different capabilities here.
264
653260
3000
大量不同的功能
10:56
Here we'll just create, in this line,
265
656260
3000
这行,我们就能建立
10:59
a little user interface that allows us to
266
659260
3000
一个简单的用户界面
11:02
do something fun there.
267
662260
2000
它允许我们做一些有趣的事情
11:05
If you go on, that's a slightly more complicated program
268
665260
2000
如果你继续的话,那就出现一些更复杂的程序
11:07
that's now doing all sorts of algorithmic things
269
667260
3000
这些程序在运行算法之类的程序
11:10
and creating user interface and so on.
270
670260
2000
并且建立用户界面等等
11:12
But it's something that is very precise stuff.
271
672260
3000
不过,这是非常精准的东西
11:15
It's a precise specification with a precise formal language
272
675260
3000
它精准的命令需要精准的正式编程语言
11:18
that causes Mathematica to know what to do here.
273
678260
3000
才能让Mathematica知道要干什么
11:21
Then on the other hand, we have Wolfram Alpha,
274
681260
3000
另一方面,我们拥有Wolfram Alpha
11:24
with all the messiness of the world
275
684260
2000
包含了世界上所有杂乱无章的东西
11:26
and human language and so on built into it.
276
686260
2000
以及人类语言等内建的知识体系
11:28
So what happens when you put these things together?
277
688260
3000
如果把他们放一起,会发生什么呢?
11:31
I think it's actually rather wonderful.
278
691260
2000
我觉得真是非常棒
11:33
With Wolfram Alpha inside Mathematica,
279
693260
2000
Mathematica里有Wolfram Alpha,
11:35
you can, for example, make precise programs
280
695260
2000
你就能编写精准的程序
11:37
that call on real world data.
281
697260
2000
来接触真实世界的数据
11:39
Here's a real simple example.
282
699260
2000
这里有个很简单的例子
11:44
You can also just sort of give vague input
283
704260
3000
你可以只是输入模棱两可的话语
11:47
and then try and have Wolfram Alpha
284
707260
2000
试着让Wolfram Alpha
11:49
figure out what you're talking about.
285
709260
2000
来分析出你想研究的内容
11:51
Let's try this here.
286
711260
2000
我们在这儿试试看
11:53
But actually I think the most exciting thing about this
287
713260
3000
不过事实上我想最激动人心的事是
11:56
is that it really gives one the chance
288
716260
2000
它给了我们一个机会
11:58
to democratize programming.
289
718260
3000
来全民编程
12:01
I mean, anyone will be able to say what they want in plain language.
290
721260
3000
我的意思是,任何人都能用日常用语说话
12:04
Then, the idea is that Wolfram Alpha will be able to figure out
291
724260
3000
关键在于,Wolfram Alpha能分析出
12:07
what precise pieces of code
292
727260
2000
什么样的精准代码
12:09
can do what they're asking for
293
729260
2000
能符合人们要求的事情
12:11
and then show them examples that will let them pick what they need
294
731260
3000
然后显示出样例来帮助人们找到想要的答案
12:14
to build up bigger and bigger, precise programs.
295
734260
3000
由此建立越来越多的精准程序
12:17
So, sometimes, Wolfram Alpha
296
737260
2000
所以,有时候,Wolfram Alpha
12:19
will be able to do the whole thing immediately
297
739260
2000
能够立即处理整个问题
12:21
and just give back a whole big program that you can then compute with.
298
741260
3000
然后仅仅回馈你能用来计算的整个大程序
12:24
Here's a big website
299
744260
2000
这里有个大网站
12:26
where we've been collecting lots of educational
300
746260
3000
这里,我们收集了很多关于教育等
12:29
and other demonstrations about lots of kinds of things.
301
749260
3000
各种事物的样例
12:32
I'll show you one example here.
302
752260
3000
我来展示一个例子,例如这个
12:36
This is just an example of one of these computable documents.
303
756260
3000
这只是可计算文档的其中一个样例
12:39
This is probably a fairly small
304
759260
2000
它是相当小的
12:41
piece of Mathematica code
305
761260
2000
一段Mathematica代码
12:43
that's able to be run here.
306
763260
2000
能在这里运行
12:47
Okay. Let's zoom out again.
307
767260
3000
我们再缩小一下
12:50
So, given our new kind of science,
308
770260
2000
所以,有了这个新版科学
12:52
is there a general way to use it to make technology?
309
772260
3000
存在一个通用的办法来用它革新技术吗?
12:55
So, with physical materials,
310
775260
2000
使用物理材料
12:57
we're used to going around the world
311
777260
2000
我们过去常常遍步世界
12:59
and discovering that particular materials
312
779260
2000
并发现特定材料
13:01
are useful for particular
313
781260
2000
用于特定的
13:03
technological purposes.
314
783260
2000
技术目的等等。
13:05
Well, it turns out we can do very much the same kind of thing
315
785260
2000
结果,我们可以做很多差不多的事情
13:07
in the computational universe.
316
787260
2000
在这个可计算的世界中。
13:09
There's an inexhaustible supply of programs out there.
317
789260
3000
有无穷无尽的程序资源在那儿。
13:12
The challenge is to see how to
318
792260
2000
面临的挑战是如何
13:14
harness them for human purposes.
319
794260
2000
让它们供人类使用
13:16
Something like Rule 30, for example,
320
796260
2000
举个例子,一些像30号规则的东西
13:18
turns out to be a really good randomness generator.
321
798260
2000
结果可以是很好的随机生成器。
13:20
Other simple programs are good models
322
800260
2000
其他简单的程序是很好的模型
13:22
for processes in the natural or social world.
323
802260
3000
来处理自然世界或者社交活动的问题
13:25
And, for example, Wolfram Alpha and Mathematica
324
805260
2000
再比如,Wolfram Alpha和Mathematica
13:27
are actually now full of algorithms
325
807260
2000
确实包含很多算法
13:29
that we discovered by searching the computational universe.
326
809260
3000
我们通过搜索计算空间找到它们
13:33
And, for example, this -- if we go back here --
327
813260
3000
再比如,我们返回到这里
13:37
this has become surprisingly popular
328
817260
2000
这个已经变成相当的流行
13:39
among composers
329
819260
2000
在作曲家间
13:41
finding musical forms by searching the computational universe.
330
821260
3000
通过搜索计算空间来找出音乐模式
13:45
In a sense, we can use the computational universe
331
825260
2000
某种意义上说,我们可以使用计算空间
13:47
to get mass customized creativity.
332
827260
3000
来获得大量的个性化创造。
13:50
I'm hoping we can, for example,
333
830260
2000
我希望我们能够
13:52
use that even to get Wolfram Alpha
334
832260
2000
使用Wolfram Alpha
13:54
to routinely do invention and discovery on the fly,
335
834260
3000
来运行常规的发明和发现的过程
13:57
and to find all sorts of wonderful stuff
336
837260
2000
并且来找出所有令人惊讶的事情
13:59
that no engineer
337
839260
2000
这些事情没有一个工程师
14:01
and no process of incremental evolution would ever come up with.
338
841260
3000
也没有一个渐进式演化的过程能够找出
14:05
Well, so, that leads to kind of an ultimate question:
339
845260
3000
这些最终导向一个终极问题
14:08
Could it be that someplace out there in the computational universe
340
848260
3000
有没有可能使这个计算空间
14:11
we might find our physical universe?
341
851260
3000
与我们的物理世界相融合?
14:14
Perhaps there's even some quite simple rule,
342
854260
2000
也许存在简单的规则
14:16
some simple program for our universe.
343
856260
3000
一些简单的程序,对于我们的物理世界来说。
14:19
Well, the history of physics would have us believe
344
859260
2000
物理的历史让我们相信
14:21
that the rule for the universe must be pretty complicated.
345
861260
3000
宇宙的内部规则一定是很复杂的
14:24
But in the computational universe,
346
864260
2000
但是在计算空间中
14:26
we've now seen how rules that are incredibly simple
347
866260
3000
我们已经看到那些规则惊人的简单
14:29
can produce incredibly rich and complex behavior.
348
869260
3000
却能够产生非常丰富和复杂的结果
14:32
So could that be what's going on with our whole universe?
349
872260
3000
所以,这可能是我们的物理世界的本质吗?
14:36
If the rules for the universe are simple,
350
876260
2000
如果这个宇宙的规则很简单
14:38
it's kind of inevitable that they have to be
351
878260
2000
不可避免的,他们一定是
14:40
very abstract and very low level;
352
880260
2000
十分抽象以及初级
14:42
operating, for example, far below
353
882260
2000
远远运行于
14:44
the level of space or time,
354
884260
2000
时间、空间之下
14:46
which makes it hard to represent things.
355
886260
2000
这种运行方法很难表现某种东西
14:48
But in at least a large class of cases,
356
888260
2000
但是至少,从其中一类大量的事例中
14:50
one can think of the universe as being
357
890260
2000
我们能把这个宇宙想成
14:52
like some kind of network,
358
892260
2000
某种网络
14:54
which, when it gets big enough,
359
894260
2000
当它变得足够大时
14:56
behaves like continuous space
360
896260
2000
它表现得像一个连续空间
14:58
in much the same way as having lots of molecules
361
898260
2000
某种程度上就像很多分子
15:00
can behave like a continuous fluid.
362
900260
2000
表现得像流体一样。
15:02
Well, then the universe has to evolve by applying
363
902260
3000
之后,宇宙进化就要依靠
15:05
little rules that progressively update this network.
364
905260
3000
应用这个网络中不断更新的简单规则。
15:08
And each possible rule, in a sense,
365
908260
2000
并且,每一个可能的规则,在某种程度上说,
15:10
corresponds to a candidate universe.
366
910260
2000
对应一个候选空间
15:12
Actually, I haven't shown these before,
367
912260
3000
事实上,我之前从来没有展示过
15:16
but here are a few of the candidate universes
368
916260
3000
不过,这里有几个候选空间
15:19
that I've looked at.
369
919260
2000
我正在研究的
15:21
Some of these are hopeless universes,
370
921260
2000
一些是没希望的空间
15:23
completely sterile,
371
923260
2000
完全不能演化,
15:25
with other kinds of pathologies like no notion of space,
372
925260
2000
包括很多缺点,例如没有空间的观念
15:27
no notion of time, no matter,
373
927260
3000
没有时间的概念,没有物质
15:30
other problems like that.
374
930260
2000
或者类似的其他问题
15:32
But the exciting thing that I've found in the last few years
375
932260
3000
但是,我近几年发现的最令人激动的事
15:35
is that you actually don't have to go very far
376
935260
2000
是你其实不必深入
15:37
in the computational universe
377
937260
2000
在计算空间中
15:39
before you start finding candidate universes
378
939260
2000
你就能发现与我们的物理空间
15:41
that aren't obviously not our universe.
379
941260
3000
明显不同的候选空间
15:44
Here's the problem:
380
944260
2000
问题在这里:
15:46
Any serious candidate for our universe
381
946260
3000
任何有可能的候选空间
15:49
is inevitably full of computational irreducibility.
382
949260
3000
不可避免地充满了计算不可化归性,
15:52
Which means that it is irreducibly difficult
383
952260
3000
这意味着简化它的具体表现
15:55
to find out how it will really behave,
384
955260
2000
是极其困难的
15:57
and whether it matches our physical universe.
385
957260
3000
并且不易判断它是否符合我们的物理世界。
16:01
A few years ago, I was pretty excited to discover
386
961260
3000
几年前,我非常兴奋地发现
16:04
that there are candidate universes with incredibly simple rules
387
964260
3000
有些候选空间具有极其简单的规则
16:07
that successfully reproduce special relativity,
388
967260
2000
却能成功再现狭义相对论
16:09
and even general relativity and gravitation,
389
969260
3000
和广义相对论以及重力
16:12
and at least give hints of quantum mechanics.
390
972260
3000
而且至少还给出了量子力学的暗示。
16:15
So, will we find the whole of physics?
391
975260
2000
所以,我们将会发现整个物理学吗?
16:17
I don't know for sure,
392
977260
2000
我不确定。
16:19
but I think at this point it's sort of
393
979260
2000
但是我觉得现在
16:21
almost embarrassing not to at least try.
394
981260
2000
不去尝试的话真的是令人羞愧的。
16:23
Not an easy project.
395
983260
2000
虽然这不是件简单的事。
16:25
One's got to build a lot of technology.
396
985260
2000
一方面要发展技术
16:27
One's got to build a structure that's probably
397
987260
2000
一方面要建立架构
16:29
at least as deep as existing physics.
398
989260
2000
这架构至少要达到现有物理学的深度。
16:31
And I'm not sure what the best way to organize the whole thing is.
399
991260
3000
而且,我不确定去整合整件事情最好的方法是什么。
16:34
Build a team, open it up, offer prizes and so on.
400
994260
3000
建立一个团队,运营它,还是提供奖励等等。
16:37
But I'll tell you, here today,
401
997260
2000
但是,我今天要告诉你
16:39
that I'm committed to seeing this project done,
402
999260
2000
我要把这个项目做完,
16:41
to see if, within this decade,
403
1001260
3000
要看看在这10年内
16:44
we can finally hold in our hands
404
1004260
2000
我们是否最终可以掌握
16:46
the rule for our universe
405
1006260
2000
我们宇宙的规则
16:48
and know where our universe lies
406
1008260
2000
并且知道我们宇宙在
16:50
in the space of all possible universes ...
407
1010260
2000
所有可能的宇宙空间的位置
16:52
and be able to type into Wolfram Alpha, "the theory of the universe,"
408
1012260
3000
并且,能够在Wolfram Alpha中输入“宇宙理论”
16:55
and have it tell us.
409
1015260
2000
让它告诉我们结果。
16:57
(Laughter)
410
1017260
2000
(笑声)
17:00
So I've been working on the idea of computation
411
1020260
2000
我已经在计算的这个想法上做了
17:02
now for more than 30 years,
412
1022260
2000
超过30年了研究
17:04
building tools and methods and turning intellectual ideas
413
1024260
3000
打造工具,创立方法,将专业知识
17:07
into millions of lines of code
414
1027260
2000
编写成数百万行的代码
17:09
and grist for server farms and so on.
415
1029260
2000
在服务器中收获结果等等。
17:11
With every passing year,
416
1031260
2000
每过去一年
17:13
I realize how much more powerful
417
1033260
2000
我都意识到
17:15
the idea of computation really is.
418
1035260
2000
计算的想法是多么的强大。
17:17
It's taken us a long way already,
419
1037260
2000
它已引领我们走过很长一段路
17:19
but there's so much more to come.
420
1039260
2000
但是还有更多可以做的事情。
17:21
From the foundations of science
421
1041260
2000
从科学的根基
17:23
to the limits of technology
422
1043260
2000
到技术的极限
17:25
to the very definition of the human condition,
423
1045260
2000
再到人类条件的定义,
17:27
I think computation is destined to be
424
1047260
2000
我觉得,计算注定
17:29
the defining idea of our future.
425
1049260
2000
是定义我们的未来的想法
17:31
Thank you.
426
1051260
2000
谢谢。
17:33
(Applause)
427
1053260
14000
(鼓掌)
17:47
Chris Anderson: That was astonishing.
428
1067260
2000
Chris Anderson(克里斯 安德森):太令人惊讶了。
17:49
Stay here. I've got a question.
429
1069260
2000
别走,我有问题。
17:51
(Applause)
430
1071260
4000
(鼓掌)
17:57
So, that was, fair to say, an astonishing talk.
431
1077260
3000
说实在的,那真的是很惊人的演讲。
18:01
Are you able to say in a sentence or two
432
1081260
3000
您能用一两句话概括
18:04
how this type of thinking
433
1084260
3000
这种思考方式如何
18:07
could integrate at some point
434
1087260
2000
能在某些点上整合
18:09
to things like string theory or the kind of things that people think of
435
1089260
2000
一些如弦论或者
18:11
as the fundamental explanations of the universe?
436
1091260
3000
人们在思考的一些关于根本宇宙解释的问题?
18:14
Stephen Wolfram: Well, the parts of physics
437
1094260
2000
Stephen Wolfram(斯蒂芬.沃尔夫勒姆):好的。
18:16
that we kind of know to be true,
438
1096260
2000
那部分我们视作真理的物理学
18:18
things like the standard model of physics:
439
1098260
2000
就像标准物理模型
18:20
what I'm trying to do better reproduce the standard model of physics
440
1100260
3000
我尝试做得更好的是再现标准物理模型
18:23
or it's simply wrong.
441
1103260
2000
或者说明它是错的。
18:25
The things that people have tried to do in the last 25 years or so
442
1105260
2000
人们在近25年里已尝试的事情
18:27
with string theory and so on
443
1107260
2000
有关弦论等等
18:29
have been an interesting exploration
444
1109260
2000
都是非常有趣的探索
18:31
that has tried to get back to the standard model,
445
1111260
3000
这些探索已经尝试回到标准模型,
18:34
but hasn't quite gotten there.
446
1114260
2000
却还不能到那一步。
18:36
My guess is that some great simplifications of what I'm doing
447
1116260
3000
我猜我的研究中的一些极端简化
18:39
may actually have considerable resonance
448
1119260
3000
可能和弦论中的某些研究
18:42
with what's been done in string theory,
449
1122260
2000
有相当的相似度
18:44
but that's a complicated math thing
450
1124260
3000
不过,那是复杂的数学
18:47
that I don't yet know how it's going to work out.
451
1127260
3000
我还不知道有些是怎么回事情。
18:50
CA: Benoit Mandelbrot is in the audience.
452
1130260
2000
克里斯 安德森: Benoit Mandlebrot也在观众席中。
18:52
He also has shown how complexity
453
1132260
2000
他也展示了如何复杂
18:54
can arise out of a simple start.
454
1134260
2000
可以从简单的初始状态演化过来。
18:56
Does your work relate to his?
455
1136260
2000
这和你的研究相关吗?
18:58
SW: I think so.
456
1138260
2000
史蒂芬:我觉得有。
19:00
I view Benoit Mandelbrot's work
457
1140260
2000
我看过Benoit Mandlebrot的研究,
19:02
as one of the founding contributions
458
1142260
3000
觉得像这个领域的
19:05
to this kind of area.
459
1145260
3000
基础贡献
19:08
Benoit has been particularly interested
460
1148260
2000
Benoit致力于
19:10
in nested patterns, in fractals and so on,
461
1150260
2000
复杂图样,分型等等的研究,
19:12
where the structure is something
462
1152260
2000
在那些方面,结构就像
19:14
that's kind of tree-like,
463
1154260
2000
树型之类的东西,
19:16
and where there's sort of a big branch that makes little branches
464
1156260
2000
有大分支,能产生小分支
19:18
and even smaller branches and so on.
465
1158260
3000
和更小分支
19:21
That's one of the ways
466
1161260
2000
那也是一种方法
19:23
that you get towards true complexity.
467
1163260
3000
来到达真正的复杂。
19:26
I think things like the Rule 30 cellular automaton
468
1166260
3000
我觉得像30号规则的单元自动机
19:29
get us to a different level.
469
1169260
2000
将我们带到了不同的水平上。
19:31
In fact, in a very precise way, they get us to a different level
470
1171260
3000
事实上,更精确地说,它能将我们带到不同的水平
19:34
because they seem to be things that are
471
1174260
2000
因为他们看似能够
19:37
capable of complexity
472
1177260
3000
达到复杂状态
19:40
that's sort of as great as complexity can ever get ...
473
1180260
3000
这种复杂是前所未有的...
19:44
I could go on about this at great length, but I won't. (Laughter) (Applause)
474
1184260
3000
我可以持续不断地讲下去,但是我不打算去做。
19:47
CA: Stephen Wolfram, thank you.
475
1187260
2000
克里斯:史蒂芬,谢谢你。
19:49
(Applause)
476
1189260
2000
(鼓掌)
关于本网站

这个网站将向你介绍对学习英语有用的YouTube视频。你将看到来自世界各地的一流教师教授的英语课程。双击每个视频页面上显示的英文字幕,即可从那里播放视频。字幕会随着视频的播放而同步滚动。如果你有任何意见或要求,请使用此联系表与我们联系。

https://forms.gle/WvT1wiN1qDtmnspy7