What intelligent machines can learn from a school of fish | Radhika Nagpal

109,666 views ・ 2017-10-06

TED


请双击下面的英文字幕来播放视频。

翻译人员: 校对人员: Yolanda Zhang
00:12
In my early days as a graduate student,
0
12575
2015
当我还在读研究生的时候
00:14
I went on a snorkeling trip off the coast of the Bahamas.
1
14614
3555
我在巴哈马海岸参加了一次潜泳之旅
00:18
I'd actually never swum in the ocean before,
2
18609
2949
我之前从没在海里游过泳
00:21
so it was a bit terrifying.
3
21582
1844
心里难免有些忐忑
00:23
What I remember the most is, as I put my head in the water
4
23836
3000
让我印象最深的是 当我把头潜入水中时
00:26
and I was trying really hard to breathe through the snorkel,
5
26860
4250
很努力地试图通过通气管呼吸
00:31
this huge group of striped yellow and black fish
6
31134
5366
这时一大群拥有黄黑相间条纹的鱼
00:36
came straight at me ...
7
36524
1483
径直向我游过来
00:38
and I just froze.
8
38637
1397
我瞬间变得不知所措
00:40
And then, as if it had suddenly changed its mind,
9
40795
3613
之后鱼群好像突然改变了主意
00:44
came towards me and then swerved to the right
10
44432
2437
游到我面前 然后突然右转
00:46
and went right around me.
11
46893
1515
与我擦肩而过
00:48
It was absolutely mesmerizing.
12
48740
1526
这鱼群彻底惊艳到我了
00:50
Maybe many of you have had this experience.
13
50290
2182
也许你们中的许多人也有过这样的经历
00:53
Of course, there's the color and the beauty of it,
14
53059
3422
当然 令人着迷的不但 有它的艳丽和壮美
00:56
but there was also just the sheer oneness of it,
15
56505
2928
同时也有它纯粹的同一性
00:59
as if it wasn't hundreds of fish
16
59457
2343
好像它不是成百上千条鱼
01:01
but a single entity with a single collective mind
17
61824
3135
而是单一 有唯一集体思维的个体
01:04
that was making decisions.
18
64983
1507
在做决定
01:07
When I look back, I think that experience really ended up determining
19
67486
3682
回想起来 这段经历 对我职业生涯的大部分工作
01:11
what I've worked on for most of my career.
20
71192
2222
起到了决定性的作用
01:15
I'm a computer scientist,
21
75977
1280
我是一个计算机科学家
01:17
and the field that I work in is artificial intelligence.
22
77281
2747
从事的领域是人工智能(AI)
01:20
And a key theme in AI
23
80459
1517
人工智能中一个关键的主题
01:22
is being able to understand intelligence by creating our own computational systems
24
82000
4443
是能通过创造我们自己的 计算机系统来了解智能
01:26
that display intelligence the way we see it in nature.
25
86467
3253
这些系统展现了我们在 自然中所观察到的智能
01:30
Now, most popular views of AI, of course, come from science fiction and the movies,
26
90287
4438
如今AI最受欢迎的想法是 源于科幻小说和电影
01:34
and I'm personally a big Star Wars fan.
27
94749
2577
我本人就是星球大战的超级粉丝
01:38
But that tends to be a very human-centric view of intelligence.
28
98321
3889
但那些关于AI的想法 往往以人类为中心
01:42
When you think of a fish school,
29
102964
2207
当你想到鱼群
01:45
or when I think of a flock of starlings,
30
105195
2953
或当我想到一群八哥时
01:48
that feels like a really different kind of intelligence.
31
108172
3340
那是一种全新的关于智能的体会
01:52
For starters, any one fish is just so tiny
32
112765
3913
首先 每一条鱼在庞大鱼群的映衬下
01:56
compared to the sheer size of the collective,
33
116702
2887
都显得非常微不足道
01:59
so it seems that any one individual
34
119613
3110
所以似乎任何一个个体
02:02
would have a really limited and myopic view of what's going on,
35
122747
2993
对于所发生事物的视角都是 非常有限和片面的
02:05
and intelligence isn't really about the individual
36
125764
2334
智能似乎与个体无关
02:08
but somehow a property of the group itself.
37
128122
2677
而是群体的某种属性
02:11
Secondly, and the thing that I still find most remarkable,
38
131938
3231
其次 我觉得最引人注目的是
02:15
is that we know that there are no leaders supervising this fish school.
39
135193
5032
鱼群中没有指挥者监督它们
02:20
Instead, this incredible collective mind behavior
40
140983
3501
这种简直不可思议的集体行为
02:24
is emerging purely from the interactions of one fish and another.
41
144508
4532
仅仅源于个体之间的互动
02:29
Somehow, there are these interactions or rules of engagement
42
149064
3968
在相邻的鱼之间以某种方式
02:33
between neighboring fish
43
153056
1755
存在着合作的规律
02:34
that make it all work out.
44
154835
1467
得以呈现这样的景象
02:37
So the question for AI then becomes,
45
157556
2651
所以AI的问题演变成了
02:40
what are those rules of engagement that lead to this kind of intelligence,
46
160231
4158
引导这种智能背后的规则是什么呢
02:44
and of course, can we create our own?
47
164413
1907
当然还包括 我们可以把它创造出来吗
02:46
And that's the primary thing that I work on with my team in my lab.
48
166819
3587
这就是我和我的团队 在实验室中的主要工作内容
02:50
We work on it through theory,
49
170763
1637
我们通过理论研究
02:52
looking at abstract rule systems
50
172424
2348
观察其中抽象的规则系统
02:54
and thinking about the mathematics behind it.
51
174796
2349
然后思考蕴藏在背后的运算规则
02:57
We also do it through biology, working closely with experimentalists.
52
177717
4285
我们也会和生物学方面的 研究人员密切合作
03:02
But mostly, we do it through robotics,
53
182399
1953
但最主要的是 我们通过机器人进行研究
03:04
where we try to create our own collective systems
54
184376
3904
尝试着去创造 属于我们自己的群体系统
03:08
that can do the kinds of things that we see in nature,
55
188304
2707
使其可以完成我们在 自然中所观察到的现象
03:11
or at least try to.
56
191035
1237
至少可以尝试一下
03:13
One of our first robotic quests along this line
57
193727
2804
由此 我们第一批的机器人探索活动
03:16
was to create our very own colony of a thousand robots.
58
196555
4045
创造了属于我们的 由一千个机器人组成的群体
03:20
So very simple robots,
59
200960
1334
非常简单的机器人
03:22
but they could be programmed to exhibit collective intelligence,
60
202318
3603
但可以通过编程让它们展现集体智慧
03:25
and that's what we were able to do.
61
205945
1729
这也是我们能够实现的
03:28
So this is what a single robot looks like.
62
208014
2032
这是单个的机器人
03:30
It's quite small, about the size of a quarter,
63
210070
2523
它非常小 只有一枚硬币大小
03:32
and you can program how it moves,
64
212617
2310
你可以通过编程控制它的移动
03:34
but it can also wirelessly communicate with other robots,
65
214951
3416
但它也可以和其他的机器人无线连接
03:38
and it can measure distances from them.
66
218391
2167
同时也可以测量它们之间的距离
03:40
And so now we can start to program exactly an interaction,
67
220582
3476
现在我们可以开始 精确地编码一个交互运动
03:44
a rule of engagement between neighbors.
68
224082
2094
两个相邻机器人之间得运动规则
03:46
And once we have this system,
69
226533
1894
一旦我们有了这个系统
03:48
we can start to program many different kinds of rules of engagement
70
228451
3416
我们就能开始编码你可以 在自然界中观察到的
03:51
that you would see in nature.
71
231891
1506
许多不同种类的规则
03:53
So for example, spontaneous synchronization,
72
233421
2976
例如自发同步现象
03:56
how audiences are clapping and suddenly start all clapping together,
73
236421
5238
观众是怎样鼓掌 怎样突然一起开始鼓掌的
04:01
the fireflies flashing together.
74
241683
2068
萤火虫如何同步闪烁
04:06
We can program rules for pattern formation,
75
246739
2691
我们可以编码图案形成的规则
04:09
how cells in a tissue
76
249454
1786
组织中的细胞
04:11
determine what role they're going to take on
77
251264
2102
如何决定它们将会起什么样的作用
04:13
and set the patterns of our bodies.
78
253390
1706
进而决定我们的人体结构的
04:16
We can program rules for migration,
79
256865
2089
我们可以编码迁移的规则
04:18
and in this way, we're really learning from nature's rules.
80
258978
2977
这样一来我们真的是在借鉴自然法则
04:22
But we can also take it a step further.
81
262415
2647
但我们也可以更进一步
04:25
We can actually take these rules that we've learned from nature
82
265086
2992
我们可以将这些从自然中学到的规律
04:28
and combine them and create entirely new collective behaviors
83
268102
3794
结合在一起 创造全新的 属于我们自己的
04:31
of our very own.
84
271920
1198
集体行为
04:33
So for example,
85
273780
1478
例如
04:35
imagine that you had two different kinds of rules.
86
275282
2352
想象一下你有两种不同的规则
04:38
So your first rule is a motion rule
87
278194
2119
第一种是运动规则
04:40
where a moving robot can move around other stationary robots.
88
280337
4341
运动的机器人可以在 其他固定的机器人周围移动
04:44
And your second rule is a pattern rule
89
284702
1811
第二种是样式规则
04:46
where a robot takes on a color based on its two nearest neighbors.
90
286537
3157
机器人参考相邻的两个机器人 呈现与其不同的颜色
04:50
So if I start with a blob of robots in a little pattern seed,
91
290499
3445
如果从一小组机器人开始
04:53
it turns out that these two rules are sufficient for the group
92
293968
2906
结果显示这两种规则已经 足够实现让群体自我组装成
04:56
to be able to self-assemble a simple line pattern.
93
296898
2752
一个简单的线性图案这一目标
05:00
And if I have more complicated pattern rules,
94
300934
2544
如果我用更复杂的样式规则
05:03
and I design error correction rules,
95
303502
2317
并且设计错误更正规则
05:05
we can actually create really, really complicated self assemblies,
96
305843
3097
我们就可以创造 非常复杂的自组装结构了
05:08
and here's what that looks like.
97
308964
1644
给大家展示一下初步成果
05:11
So here, you're going to see a thousand robots
98
311694
2985
这里你们可以看到一千个机器人
05:14
that are working together to self-assemble the letter K.
99
314703
3462
合作自发组成字母K
05:18
The K is on its side.
100
318189
1306
完成后的字母位于上方
05:20
And the important thing is that no one is in charge.
101
320043
2731
重要的是没有一个 机器人起指导作用
05:22
So any single robot is only talking to a small number of robots nearby it,
102
322798
4825
每个机器人 只和周围的一些机器人交流
05:27
and it's using its motion rule to move around the half-built structure
103
327647
3937
它利用自身的运动规则 在建了一半的结构周围运动
05:31
just looking for a place to fit in based on its pattern rules.
104
331608
3007
只是根据其样式规律 寻找一个合适的位置
05:35
And even though no robot is doing anything perfectly,
105
335614
4398
即使没有一个机器人的行为是完美的
05:40
the rules are such that we can get the collective to do its goal
106
340036
3660
这些规则却足以使我们通过集体协作
顺利达成目标
05:43
robustly together.
107
343720
1473
05:45
And the illusion becomes almost so perfect, you know --
108
345853
2982
这种幻想近乎完美
05:48
you just start to not even notice that they're individual robots at all,
109
348859
3416
你们甚至不需要注意到 它们是个体机器人
05:52
and it becomes a single entity,
110
352299
1683
它们变成了一个单一的实体
05:54
kind of like the school of fish.
111
354006
1721
就像刚刚提到的鱼群一样
05:59
So these are robots and rules in two dimensions,
112
359833
2739
这些是从二维层面看 机器人及其运动规则
06:02
but we can also think about robots and rules in three dimensions.
113
362596
3311
但我们也可以从三维的角度 考虑机器人和规则
06:05
So what if we could create robots that could build together?
114
365931
3603
如果我们能创造 有建筑能力的机器人呢
06:10
And here, we can take inspiration from social insects.
115
370396
3255
看看这里 我们从 社会性昆虫身上获得了启发
06:14
So if you think about mound-building termites
116
374009
2660
如果你想到了白蚁丘
06:16
or you think about army ants,
117
376693
2052
或者是行军蚁
06:18
they create incredible, complex nest structures out of mud
118
378769
4253
它们用泥土 甚至用它们自己的身体
06:23
and even out of their own bodies.
119
383046
2144
惊人 复杂的巢穴结构
06:26
And like the system I showed you before,
120
386422
2220
就像我之前给你们展示的系统一样
06:28
these insects actually also have pattern rules
121
388666
2970
这些昆虫实际上也有排布规则
06:31
that help them determine what to build,
122
391660
2038
可以帮助它们决定要建造什么
06:33
but the pattern can be made out of other insects,
123
393722
2302
但是这种样式结构 也可以由其他昆虫造出来
06:36
or it could be made out of mud.
124
396048
1787
或者是用泥土
06:38
And we can use that same idea to create rules for robots.
125
398998
4361
我们也可以将同样的想法 用于创建机器人规则上
06:44
So here, you're going to see some simulated robots.
126
404041
3161
这里你们可以看到一些模拟机器人
06:47
So the simulated robot has a motion rule,
127
407226
2483
模拟机器人有自己的运动规则
06:49
which is how it traverses through the structure,
128
409733
2333
指导它如何穿越结构体
06:52
looking for a place to fit in,
129
412090
1997
找到一个合适的地方
06:54
and it has pattern rules where it looks at groups of blocks
130
414111
3000
同时它也有排布规则 它会查看组块
06:57
to decide whether to place a block.
131
417135
2205
来决定是否放下一个方块
07:00
And with the right motion rules and the right pattern rules,
132
420464
3063
有了正确的运动和样式规则之后
07:03
we can actually get the robots to build whatever we want.
133
423551
3635
我们就能让机器人建造出 我们想要的任何东西
07:08
And of course, everybody wants their own tower.
134
428017
2691
当然每个人都想要属于自己的塔楼
07:11
(Laughter)
135
431170
1982
(笑)
07:13
So once we have these rules,
136
433820
1684
一旦我们有了这些规则
07:15
we can start to create the robot bodies that go with these rules.
137
435528
3166
我们就可以开始创造 能遵守这些规则的机器人
07:18
So here, you see a robot that can climb over blocks,
138
438718
3309
在这儿你们能看到 可以爬上方块的机器人
07:22
but it can also lift and move these blocks
139
442051
2681
但它也可以举起和移动这些方块
07:24
and it can start to edit the very structure that it's on.
140
444756
2697
可以编辑它所在的结构体
07:28
But with these rules,
141
448437
1148
但即便有了这些规则
07:29
this is really only one kind of robot body that you could imagine.
142
449609
3479
这也仅仅只是你们能想到的 机器人中的一种
07:33
You could imagine many different kinds of robot bodies.
143
453112
2579
你们可以想象许多不同种类的机器人
07:35
So if you think about robots that maybe could move sandbags
144
455715
4610
如果你觉得机器人也许可以移动沙袋
07:40
and could help build levees,
145
460349
2549
并帮助建筑堤坝
07:42
or we could think of robots that built out of soft materials
146
462922
4301
或者我们想象机器人 可以用柔软的材料造成
07:47
and worked together to shore up a collapsed building --
147
467247
3644
共同支撑倒塌的建筑
07:50
so just the same kind of rules in different kinds of bodies.
148
470915
2998
只是把同样的规则 用于不同种类的机器人
07:56
Or if, like my group, you are completely obsessed with army ants,
149
476030
4223
或是像我的团队一样 你完全着迷于行军蚁
08:00
then maybe one day we can make robots that can climb over literally anything
150
480277
4374
那么也许有一天我们能 让机器人攀爬所有的东西
08:04
including other members of their tribe,
151
484675
2174
包括它们部落中的其它成员
08:06
and self-assemble things out of their own bodies.
152
486873
2349
或者用它们自身自组装一些东西
08:09
Once you understand the rules,
153
489957
1681
一旦你明白了这些规律
08:11
just many different kinds of robot visions become possible.
154
491662
3379
许多不同的机器人愿景都将成为可能
08:18
And coming back to the snorkeling trip,
155
498612
2234
再回到那个潜泳之旅
08:20
we actually understand a great deal about the rules that fish schools use.
156
500870
5345
事实上我们非常了解鱼群的运动规则
08:26
So if we can invent the bodies to go with that,
157
506589
2836
如果我们能创造一个那样的群体
08:29
then maybe there is a future
158
509449
1428
那么也许在未来
08:30
where I and my group will get to snorkel with a fish school of our own creation.
159
510901
4522
我的团队将和我们创造的鱼群 展开一次潜泳之旅
08:40
Each of these systems that I showed you
160
520670
2129
我向你们展示的每一个系统
08:42
brings us closer to having the mathematical and the conceptual tools
161
522823
4277
都使我们更接近利用数学和概念工具
08:47
to create our own versions of collective power,
162
527124
3381
来创造属于我们的集体力量这一目标
08:50
and this can enable many different kinds of future applications,
163
530529
3001
这也使许多不同的未来应用成为可能
08:53
whether you think about robots that build flood barriers
164
533554
3164
机器人可以建筑防洪堤
08:56
or you think about robotic bee colonies that could pollinate crops
165
536742
4297
可以给农作物授粉
09:01
or underwater schools of robots that monitor coral reefs,
166
541063
3524
或是让水下机器人模拟珊瑚礁
09:04
or if we reach for the stars and we thinking about programming
167
544611
3103
又或者假设我们抵达某个星球
09:07
constellations of satellites.
168
547738
1619
并设计了卫星的行星
09:09
In each of these systems,
169
549968
1612
在每一个系统中
09:11
being able to understand how to design the rules of engagement
170
551604
3547
能明白如何设计交互规则
09:15
and being able to create good collective behavior
171
555175
2514
以及创造良好的集体行为
09:17
becomes a key to realizing these visions.
172
557713
2374
成为了实现这些愿景的关键
09:22
So, so far I've talked about rules for insects and for fish
173
562562
4107
那么到目前为 止 我谈到了昆虫 鱼群
09:26
and for robots,
174
566693
2369
还有机器人的各种行为规则
09:29
but what about the rules that apply to our own human collective?
175
569086
3103
但是那些适用于人类集体的规则呢
09:32
And the last thought that I'd like to leave you with
176
572686
2430
我想留给你们的最后一点思考是
09:35
is that science is of course itself
177
575140
1681
科学本身 毫无疑问就是
09:36
an incredible manifestation of collective intelligence,
178
576845
3484
集体智慧的伟大体现
09:40
but unlike the beautiful fish schools that I study,
179
580353
3318
但不同于我所研究的美丽鱼群
09:43
I feel we still have a much longer evolutionary path to walk.
180
583695
3943
我觉得我们还有很长的一段路要走
09:48
So in addition to working on improving the science of robot collectives,
181
588566
4604
除了要致力于改进机器人集体的科学
09:53
I also work on creating robots and thinking about rules
182
593194
3277
我还致力于创造机器人和思考
09:56
that will improve our own scientific collective.
183
596495
2460
能够改善我们的科学团队的规则
10:00
There's this saying that I love:
184
600018
1668
有一句谚语我很喜欢
10:01
who does science determines what science gets done.
185
601710
3404
科学的成果掌握在从事科学的人手中
10:06
Imagine a society
186
606059
2941
想象这样一个社会
我们有交互规则
10:09
where we had rules of engagement
187
609024
1651
10:10
where every child grew up believing that they could stand here
188
610699
3303
使得每个孩子从小就 相信自己长大后能站在这里
10:14
and be a technologist of the future,
189
614026
2422
并成为未来的技术专家
10:16
or where every adult
190
616472
1501
或者每一个成年人
10:17
believed that they had the ability not just to understand but to change
191
617997
4119
都相信他们不仅有能力去理解 更有能力去改变
10:22
how science and technology impacts their everyday lives.
192
622140
3555
科技如何影响他们的日常生活
10:26
What would that society look like?
193
626640
1899
那样的社会该有多么美好啊
10:30
I believe that we can do that.
194
630206
1508
我相信我们能做得到
10:31
I believe that we can choose our rules,
195
631738
2291
我相信我们可以选择自己的规则
10:34
and we engineer not just robots
196
634053
1757
我们不仅能设计机器人
10:35
but we can engineer our own human collective,
197
635834
2596
更可以设计我们的人类群体
10:38
and if we do and when we do, it will be beautiful.
198
638454
3834
在我们开始行动时 一切都将变得非常美好
10:42
Thank you.
199
642312
1151
谢谢
10:43
(Applause)
200
643487
6547
(掌声)
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