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

109,692 views ・ 2017-10-06

TED


請雙擊下方英文字幕播放視頻。

譯者: Lilian Chiu 審譯者: Wilde Luo
00:12
In my early days as a graduate student,
0
12575
2015
在我剛開始成為研究生的時候,
00:14
I went on a snorkeling trip off the coast of the Bahamas.
1
14614
3555
我到巴哈馬海岸去浮潛。
00:18
I'd actually never swum in the ocean before,
2
18609
2949
我其實從未在海洋中游泳過,
00:21
so it was a bit terrifying.
3
21582
1844
所以我有點害怕。
00:23
What I remember the most is, as I put my head in the water
4
23836
3000
我最難忘的是,當我把頭沉入水中,
00:26
and I was trying really hard to breathe through the snorkel,
5
26860
4250
並竭力透過呼吸管呼吸,
00:31
this huge group of striped yellow and black fish
6
31134
5366
有一大群黃黑條紋的魚
00:36
came straight at me ...
7
36524
1483
筆直朝我遊來……
00:38
and I just froze.
8
38637
1397
我呆住了。
00:40
And then, as if it had suddenly changed its mind,
9
40795
3613
然後,牠們好像突然轉念了一樣,
00:44
came towards me and then swerved to the right
10
44432
2437
朝我過來之後就向右急轉彎,
00:46
and went right around me.
11
46893
1515
從我身邊繞過。
00:48
It was absolutely mesmerizing.
12
48740
1526
那實在非常迷人。
00:50
Maybe many of you have had this experience.
13
50290
2182
也許在座有許多人有過這種體驗。
00:53
Of course, there's the color and the beauty of it,
14
53059
3422
當然,魚群的顏色及美麗都很難忘,
00:56
but there was also just the sheer oneness of it,
15
56505
2928
但牠們還有著一種純粹的一體感,
00:59
as if it wasn't hundreds of fish
16
59457
2343
彷彿牠們並不是數百條魚,
01:01
but a single entity with a single collective mind
17
61824
3135
而是一個整體,包含著 一個做出決策的集體思維。
01:04
that was making decisions.
18
64983
1507
01:07
When I look back, I think that experience really ended up determining
19
67486
3682
回想起來,我認為那段經歷 使我最終下定決心
01:11
what I've worked on for most of my career.
20
71192
2222
去做這份佔據我大半生涯的工作。
01:15
I'm a computer scientist,
21
75977
1280
我是個計算機科學家,
01:17
and the field that I work in is artificial intelligence.
22
77281
2747
我研究的領域是人工智慧。
01:20
And a key theme in AI
23
80459
1517
人工智慧的關鍵主題 是要能理解「智慧」的本質,
01:22
is being able to understand intelligence by creating our own computational systems
24
82000
4443
做法是創建自己的計算系統 (computational system)
01:26
that display intelligence the way we see it in nature.
25
86467
3253
來展現類似於自然生物的智慧。
01:30
Now, most popular views of AI, of course, come from science fiction and the movies,
26
90287
4438
當然,目前最熱門的人工智慧觀點 來自科幻小說和電影,
01:34
and I'm personally a big Star Wars fan.
27
94749
2577
我個人是《星際大戰》的忠實粉絲。
01:38
But that tends to be a very human-centric view of intelligence.
28
98321
3889
但那往往是個非常 以人為中心的智慧觀。
01:42
When you think of a fish school,
29
102964
2207
當你思考魚群
01:45
or when I think of a flock of starlings,
30
105195
2953
或想像一群椋鳥,
01:48
that feels like a really different kind of intelligence.
31
108172
3340
那感覺是一種完全 不同的智慧形式。
01:52
For starters, any one fish is just so tiny
32
112765
3913
首先,和整體魚群的大小相比較,
01:56
compared to the sheer size of the collective,
33
116702
2887
一條魚真的是太小了,
01:59
so it seems that any one individual
34
119613
3110
所以,似乎其中任何一個個體
02:02
would have a really limited and myopic view of what's going on,
35
122747
2993
對正在發生的事應該 眼光短淺、缺乏遠見。
02:05
and intelligence isn't really about the individual
36
125764
2334
而且「智慧」並不體現在個體身上,
02:08
but somehow a property of the group itself.
37
128122
2677
而是團體本身的一種特性。
02:11
Secondly, and the thing that I still find most remarkable,
38
131938
3231
第二,我仍然認為是最了不起的事,
02:15
is that we know that there are no leaders supervising this fish school.
39
135193
5032
就是我們知道在這魚群中 並不存在管理著群體的領導者。
02:20
Instead, this incredible collective mind behavior
40
140983
3501
反而,這個集體思維 所做出的非凡行為
02:24
is emerging purely from the interactions of one fish and another.
41
144508
4532
單純來自魚與魚間的互動。
02:29
Somehow, there are these interactions or rules of engagement
42
149064
3968
不知何故,相鄰近的魚之間 會存在著這些互動,
02:33
between neighboring fish
43
153056
1755
或者說是約定好的行為規則,
02:34
that make it all work out.
44
154835
1467
從而產生這集體行為。
02:37
So the question for AI then becomes,
45
157556
2651
所以,對人工智慧的問題變成是:
02:40
what are those rules of engagement that lead to this kind of intelligence,
46
160231
4158
是什麼約定規則產生這種智慧的?
02:44
and of course, can we create our own?
47
164413
1907
當然還有,我們能否自己創造一個?
02:46
And that's the primary thing that I work on with my team in my lab.
48
166819
3587
這是我與團隊的實驗研究主題。
02:50
We work on it through theory,
49
170763
1637
我們透過理論來研究,
02:52
looking at abstract rule systems
50
172424
2348
探究抽象的規則系統,
02:54
and thinking about the mathematics behind it.
51
174796
2349
思考其背後的數學原理。
02:57
We also do it through biology, working closely with experimentalists.
52
177717
4285
我們也透過生物學來研究,
與實驗者密切合作。
03:02
But mostly, we do it through robotics,
53
182399
1953
但最主要是通過機器人研究,
03:04
where we try to create our own collective systems
54
184376
3904
嘗試創造我們自己的集體系統,
03:08
that can do the kinds of things that we see in nature,
55
188304
2707
讓系統能做出,或至少試著做出 自然界中的智慧行為。
03:11
or at least try to.
56
191035
1237
03:13
One of our first robotic quests along this line
57
193727
2804
我們最初以這種方式 在機器人方面的探索之一,
03:16
was to create our very own colony of a thousand robots.
58
196555
4045
是創造我們自己的千人機器人群體。
03:20
So very simple robots,
59
200960
1334
機器人非常簡單,
03:22
but they could be programmed to exhibit collective intelligence,
60
202318
3603
但能通過程式設計讓它們 展現出集體智慧,
03:25
and that's what we were able to do.
61
205945
1729
這是我們能夠做到的。
03:28
So this is what a single robot looks like.
62
208014
2032
單個的機器人看起來是這樣的。
03:30
It's quite small, about the size of a quarter,
63
210070
2523
它很小,約 25 分硬幣的大小,
03:32
and you can program how it moves,
64
212617
2310
你可以設計程式來規範它如何移動,
03:34
but it can also wirelessly communicate with other robots,
65
214951
3416
它也能以無線的方式 和其他機器人溝通,
03:38
and it can measure distances from them.
66
218391
2167
能測量與其他機器人的距離。
03:40
And so now we can start to program exactly an interaction,
67
220582
3476
我們就可以開始 針對一套互動規則來設計程式,
03:44
a rule of engagement between neighbors.
68
224082
2094
指定鄰近機器人之間的行為規則。
03:46
And once we have this system,
69
226533
1894
一旦有了這個系統,
03:48
we can start to program many different kinds of rules of engagement
70
228451
3416
我們就可針對自然界中的 各類約定規則來編寫程式。
03:51
that you would see in nature.
71
231891
1506
03:53
So for example, spontaneous synchronization,
72
233421
2976
比如「自發性同步」,
03:56
how audiences are clapping and suddenly start all clapping together,
73
236421
5238
一旦有觀眾開始拍手, 全部都驟然跟著拍手,
04:01
the fireflies flashing together.
74
241683
2068
螢火蟲也會一起發光。
04:06
We can program rules for pattern formation,
75
246739
2691
我們可以編寫圖案形成的規則, (pattern formation)
04:09
how cells in a tissue
76
249454
1786
組織中的細胞
04:11
determine what role they're going to take on
77
251264
2102
如何決定它們將扮演什麼角色
04:13
and set the patterns of our bodies.
78
253390
1706
並設定我們身體的模式。
04:16
We can program rules for migration,
79
256865
2089
我們可編寫遷移的規則,
04:18
and in this way, we're really learning from nature's rules.
80
258978
2977
以這種方式,我們能真正地 向自然界的規則學習。
04:22
But we can also take it a step further.
81
262415
2647
但,我們也可以再進一步。
04:25
We can actually take these rules that we've learned from nature
82
265086
2992
我們可以組合這些 向自然界學來的規則,
04:28
and combine them and create entirely new collective behaviors
83
268102
3794
創造出我們自己的、 全新的集體行為。
04:31
of our very own.
84
271920
1198
04:33
So for example,
85
273780
1478
比如,
04:35
imagine that you had two different kinds of rules.
86
275282
2352
想像你有兩種不同的規則。
04:38
So your first rule is a motion rule
87
278194
2119
第一種是動作規則,
04:40
where a moving robot can move around other stationary robots.
88
280337
4341
讓移動中的機器人 可以繞著靜止的機器人轉動。
04:44
And your second rule is a pattern rule
89
284702
1811
第二種是模式規則,
04:46
where a robot takes on a color based on its two nearest neighbors.
90
286537
3157
機器人會根據旁邊 兩名同伴的顔色來呈現顏色。
04:50
So if I start with a blob of robots in a little pattern seed,
91
290499
3445
所以,最開始我只需一小群機器人,
就能埋下一顆「模式種子」,
04:53
it turns out that these two rules are sufficient for the group
92
293968
2906
結果,對這個群體而言,
有這兩種規則就足以自我組裝出
04:56
to be able to self-assemble a simple line pattern.
93
296898
2752
一個簡單的線條樣式。
05:00
And if I have more complicated pattern rules,
94
300934
2544
如果我有更複雜的模式規則
05:03
and I design error correction rules,
95
303502
2317
且設計出修正錯誤的規則,
05:05
we can actually create really, really complicated self assemblies,
96
305843
3097
我們就能實際造出 非常複雜的自我組裝樣式,
05:08
and here's what that looks like.
97
308964
1644
看起來就會像是這樣。
05:11
So here, you're going to see a thousand robots
98
311694
2985
所以,各位將會在這裡 看到一千個機器人,
05:14
that are working together to self-assemble the letter K.
99
314703
3462
它們正在合作並自我組裝出 英文字母「K」。
05:18
The K is on its side.
100
318189
1306
這是一個側過來的 K 。
05:20
And the important thing is that no one is in charge.
101
320043
2731
重要的是,沒有人在主導。
05:22
So any single robot is only talking to a small number of robots nearby it,
102
322798
4825
所以任何一個機器人都只是在 和它附近的少數幾個機器人交談,
05:27
and it's using its motion rule to move around the half-built structure
103
327647
3937
它會用它的動作規則, 在這個半成品周圍移動,
05:31
just looking for a place to fit in based on its pattern rules.
104
331608
3007
根據它的模式規則, 找個適合的位置插進去。
05:35
And even though no robot is doing anything perfectly,
105
335614
4398
雖然沒有任一機器人 完美地做好一件事,
05:40
the rules are such that we can get the collective to do its goal
106
340036
3660
規則是這樣的,
我們可以讓集體一起 穩健地完成目標。
05:43
robustly together.
107
343720
1473
05:45
And the illusion becomes almost so perfect, you know --
108
345853
2982
這個幻覺幾乎完美,
05:48
you just start to not even notice that they're individual robots at all,
109
348859
3416
你甚至會忘了它們各自是個機器人,
05:52
and it becomes a single entity,
110
352299
1683
合起來成了單一的實體,
05:54
kind of like the school of fish.
111
354006
1721
就像一群魚。
05:59
So these are robots and rules in two dimensions,
112
359833
2739
上面這些是二維世界中的 機器人及規則,
06:02
but we can also think about robots and rules in three dimensions.
113
362596
3311
但我們也可以思考 三維世界中的機器人及規則。
06:05
So what if we could create robots that could build together?
114
365931
3603
如果我們造出能 共同建設的機器人會如何呢?
06:10
And here, we can take inspiration from social insects.
115
370396
3255
這裡,我們的靈感來自於群居昆蟲。
06:14
So if you think about mound-building termites
116
374009
2660
如果你想到建立土墩的白蟻
06:16
or you think about army ants,
117
376693
2052
或是行軍蟻,
06:18
they create incredible, complex nest structures out of mud
118
378769
4253
牠們造出很了不起、 很複雜的巢穴結構,
用泥巴,甚至用自己的身體。
06:23
and even out of their own bodies.
119
383046
2144
06:26
And like the system I showed you before,
120
386422
2220
就像我先前給各位看的系統,
06:28
these insects actually also have pattern rules
121
388666
2970
這些昆蟲其實也有模式規則
06:31
that help them determine what to build,
122
391660
2038
來協助牠們決定要建造什麼,
06:33
but the pattern can be made out of other insects,
123
393722
2302
做模型的材料可以是其他昆蟲
06:36
or it could be made out of mud.
124
396048
1787
甚至是泥巴。
06:38
And we can use that same idea to create rules for robots.
125
398998
4361
我們可以把同樣的想法 用來為機器人創造規則。
06:44
So here, you're going to see some simulated robots.
126
404041
3161
在這裡你將看到的 是一些模擬的機器人。
06:47
So the simulated robot has a motion rule,
127
407226
2483
這模擬機器人有一條動作規則:
06:49
which is how it traverses through the structure,
128
409733
2333
以何種方式在結構中來回移動,
06:52
looking for a place to fit in,
129
412090
1997
並尋找一個適合插入的地方。
06:54
and it has pattern rules where it looks at groups of blocks
130
414111
3000
同樣它也有一套模式規則,
使它在看到一堆積木時 決定是否放下手中的積木。
06:57
to decide whether to place a block.
131
417135
2205
07:00
And with the right motion rules and the right pattern rules,
132
420464
3063
有正確的動作規則 和正確的模式規則,
07:03
we can actually get the robots to build whatever we want.
133
423551
3635
我們就能夠讓機器人建造出 任何我們想要的東西。
07:08
And of course, everybody wants their own tower.
134
428017
2691
當然,每個人都想擁有 屬於自己的一座塔。
07:11
(Laughter)
135
431170
1982
(笑聲)
07:13
So once we have these rules,
136
433820
1684
一旦我們有了這些規則,
07:15
we can start to create the robot bodies that go with these rules.
137
435528
3166
我們就可以配合這些規則 開始打造機器人的身體。
07:18
So here, you see a robot that can climb over blocks,
138
438718
3309
在這裡,各位可以看到, 機器人能爬過積木,
07:22
but it can also lift and move these blocks
139
442051
2681
它也可以舉起和搬動這些積木,
07:24
and it can start to edit the very structure that it's on.
140
444756
2697
它可以自己開始修建這個結構。
07:28
But with these rules,
141
448437
1148
但是配合這些規則,
07:29
this is really only one kind of robot body that you could imagine.
142
449609
3479
這其實只是所有你能想到的 機器人身體構造情況中的一種。
07:33
You could imagine many different kinds of robot bodies.
143
453112
2579
你還可想像出多種 不同的機器人身體構造。
07:35
So if you think about robots that maybe could move sandbags
144
455715
4610
所以,你也許可以想像出 會搬移沙袋的機器人,
07:40
and could help build levees,
145
460349
2549
它們能協助築堤,
07:42
or we could think of robots that built out of soft materials
146
462922
4301
我們或許也可用軟材料做機器人,
07:47
and worked together to shore up a collapsed building --
147
467247
3644
共同撐起倒塌的建築物。
07:50
so just the same kind of rules in different kinds of bodies.
148
470915
2998
這只是把同樣的規則 放到不同類的身體中。
07:56
Or if, like my group, you are completely obsessed with army ants,
149
476030
4223
或者,和我的團隊一樣, 你可能對行軍蟻很著迷,
08:00
then maybe one day we can make robots that can climb over literally anything
150
480277
4374
那麼也許有一天
我們做出能爬過任何東西的機器人,
08:04
including other members of their tribe,
151
484675
2174
包括爬過它們自己的夥伴成員,
08:06
and self-assemble things out of their own bodies.
152
486873
2349
用它們自己的身體組裝出東西。
08:09
Once you understand the rules,
153
489957
1681
一旦你瞭解了規則,
08:11
just many different kinds of robot visions become possible.
154
491662
3379
多種不同類型的 機器人遠景都變為可能。
08:18
And coming back to the snorkeling trip,
155
498612
2234
回到我的浮潛之旅,
08:20
we actually understand a great deal about the rules that fish schools use.
156
500870
5345
其實我們瞭解很多魚群的規則。
08:26
So if we can invent the bodies to go with that,
157
506589
2836
所以,若我們能發明出 配合這些規則的身體,
08:29
then maybe there is a future
158
509449
1428
那麼也許在未來,
08:30
where I and my group will get to snorkel with a fish school of our own creation.
159
510901
4522
我和團隊會和我們創造出的 魚群一起浮潛。
08:40
Each of these systems that I showed you
160
520670
2129
每一個我展現給你們的系統
08:42
brings us closer to having the mathematical and the conceptual tools
161
522823
4277
讓我們更進一步邁向 這些數學和概念性工具
08:47
to create our own versions of collective power,
162
527124
3381
來創造我們自己的集體力量,
08:50
and this can enable many different kinds of future applications,
163
530529
3001
這就能讓許多種 未來技術都成為可能,
08:53
whether you think about robots that build flood barriers
164
533554
3164
你可考慮用機器人來建立防洪設施,
08:56
or you think about robotic bee colonies that could pollinate crops
165
536742
4297
用機器蜜蜂群來授粉,
09:01
or underwater schools of robots that monitor coral reefs,
166
541063
3524
或用水底機器人群體來監看珊瑚礁;
09:04
or if we reach for the stars and we thinking about programming
167
544611
3103
或是我們雄心萬丈,
可以考慮為一群衛星設計程式。
09:07
constellations of satellites.
168
547738
1619
09:09
In each of these systems,
169
549968
1612
在所有這些系統中,
09:11
being able to understand how to design the rules of engagement
170
551604
3547
能夠瞭解如何設計出約定規則,
09:15
and being able to create good collective behavior
171
555175
2514
以及能夠創造出好的集體行為,
09:17
becomes a key to realizing these visions.
172
557713
2374
是實現這些遠景的關鍵。
09:22
So, so far I've talked about rules for insects and for fish
173
562562
4107
目前,我已經談過了昆蟲、魚
09:26
and for robots,
174
566693
2369
和機器人之間的規則,
09:29
but what about the rules that apply to our own human collective?
175
569086
3103
那麼用在我們自己 人類群體上的規則呢?
09:32
And the last thought that I'd like to leave you with
176
572686
2430
最後我想留給各位 去思考的一件事是
09:35
is that science is of course itself
177
575140
1681
當然科學本身是
09:36
an incredible manifestation of collective intelligence,
178
576845
3484
集體智慧的一種偉大表現形式,
09:40
but unlike the beautiful fish schools that I study,
179
580353
3318
但不像我研究的美麗魚群,
09:43
I feel we still have a much longer evolutionary path to walk.
180
583695
3943
我覺得我們還有 非常長的演化之路要走。
09:48
So in addition to working on improving the science of robot collectives,
181
588566
4604
所以除了致力於發展機器人 群體的科學研究之外,
09:53
I also work on creating robots and thinking about rules
182
593194
3277
我也從事創造機器人的工作, 並且思考一些規則,
09:56
that will improve our own scientific collective.
183
596495
2460
它將對我們自己的 科學研究群體大有裨益。
10:00
There's this saying that I love:
184
600018
1668
分享一句我喜歡的話:
10:01
who does science determines what science gets done.
185
601710
3404
做科學的人,決定了科學能做什麽。
10:06
Imagine a society
186
606059
2941
想像一個這樣的社會:
10:09
where we had rules of engagement
187
609024
1651
我們有個約定規則:
10:10
where every child grew up believing that they could stand here
188
610699
3303
每個孩子在成長的過程中都相信
他們能站在這個講臺上
10:14
and be a technologist of the future,
189
614026
2422
成為未來的科技專家;
10:16
or where every adult
190
616472
1501
或每個成年人都相信他們有能力
10:17
believed that they had the ability not just to understand but to change
191
617997
4119
不僅理解而且改變 科技對日常生活的影響。
10:22
how science and technology impacts their everyday lives.
192
622140
3555
10:26
What would that society look like?
193
626640
1899
那樣的社會會是怎樣的?
10:30
I believe that we can do that.
194
630206
1508
我相信我們能讓它成真。
10:31
I believe that we can choose our rules,
195
631738
2291
我相信我們能選擇我們的規則,
10:34
and we engineer not just robots
196
634053
1757
除了機器人之外,
10:35
but we can engineer our own human collective,
197
635834
2596
我們也能設計我們自己的人類群體,
10:38
and if we do and when we do, it will be beautiful.
198
638454
3834
如果我們做到了, 世界會變得無比美好。
10:42
Thank you.
199
642312
1151
謝謝。
10:43
(Applause)
200
643487
6547
(掌聲)
關於本網站

本網站將向您介紹對學習英語有用的 YouTube 視頻。 您將看到來自世界各地的一流教師教授的英語課程。 雙擊每個視頻頁面上顯示的英文字幕,從那裡播放視頻。 字幕與視頻播放同步滾動。 如果您有任何意見或要求,請使用此聯繫表與我們聯繫。

https://forms.gle/WvT1wiN1qDtmnspy7