TEDxBrussels
Jeremy Howard: The wonderful and terrifying implications of computers that can learn
杰里米 霍华德: 会学习的电脑带来的美好和恐怖
Filmed:
Readability: 4.5
2,532,971 views
如果我们教电脑学习会发生什么？技术专家杰里米 霍华德分享了一些快速发展的深度学习领域里令人惊奇的新发展，让计算机能够学习汉语，识别物体和图片，或者帮助医疗诊断。（一个深度学习工具在看了一个小时YouTube视频后，教会自己“猫”这个概念。）深度学习会改变你身边的电脑得行为...远比你想象的要快。
Jeremy Howard - Data scientist
Jeremy Howard imagines how advanced machine learning can improve our lives. Full bio
Jeremy Howard imagines how advanced machine learning can improve our lives. Full bio
Double-click the English transcript below to play the video.
00:12
It used to be that if you wanted
to get a computer to do something new,
to get a computer to do something new,
0
880
4013
在过去，如果你想让计算机做一件事
00:16
you would have to program it.
1
4893
1554
你需要设计电脑程序
00:18
Now, programming, for those of you here
that haven't done it yourself,
that haven't done it yourself,
2
6447
3411
你们可能从没做过这件事
00:21
requires laying out in excruciating detail
3
9858
3502
编程需要排列出你想让电脑做的
每一个细枝末节的小步骤来达到你的目的
每一个细枝末节的小步骤来达到你的目的
00:25
every single step that you want
the computer to do
the computer to do
4
13360
3367
00:28
in order to achieve your goal.
5
16727
2362
00:31
Now, if you want to do something
that you don't know how to do yourself,
that you don't know how to do yourself,
6
19089
3496
假如你自己都不清楚完成这某件事的话
00:34
then this is going
to be a great challenge.
to be a great challenge.
7
22585
2063
要编写处电脑程序来完成那件事就会显得
比登天还要困难
比登天还要困难
00:36
So this was the challenge faced
by this man, Arthur Samuel.
by this man, Arthur Samuel.
8
24648
3483
这也是这个人，亚瑟 塞缪尔，所面临的挑战
00:40
In 1956, he wanted to get this computer
9
28131
4077
在1956年，他想让这台电脑和他下国际象棋
00:44
to be able to beat him at checkers.
10
32208
2340
00:46
How can you write a program,
11
34548
2040
你怎样才能罗列出所有的细枝末节，
并且让电脑下象棋比你厉害？
并且让电脑下象棋比你厉害？
00:48
lay out in excruciating detail,
how to be better than you at checkers?
how to be better than you at checkers?
12
36588
3806
00:52
So he came up with an idea:
13
40394
1722
他想出一个办法
00:54
he had the computer play
against itself thousands of times
against itself thousands of times
14
42116
3724
它让电脑和自己对战几千次
00:57
and learn how to play checkers.
15
45840
2524
学习如何下象棋
01:00
And indeed it worked,
and in fact, by 1962,
and in fact, by 1962,
16
48364
3180
事实证明他做到了。1962年
01:03
this computer had beaten
the Connecticut state champion.
the Connecticut state champion.
17
51544
4017
这台电脑打败了美国康涅狄克州象棋冠军
01:07
So Arthur Samuel was
the father of machine learning,
the father of machine learning,
18
55561
2973
亚瑟 塞缪尔是机器学习之父
01:10
and I have a great debt to him,
19
58534
1717
我非常敬畏他
01:12
because I am a machine
learning practitioner.
learning practitioner.
20
60251
2763
因为我是机器学习的实践者
01:15
I was the president of Kaggle,
21
63014
1465
我曾是Kaggle的主席
01:16
a community of over 200,000
machine learning practictioners.
machine learning practictioners.
22
64479
3388
Kaggle是一个拥有200,000机器学习实践者地社区
01:19
Kaggle puts up competitions
23
67867
2058
Kaggle会组织竞赛
01:21
to try and get them to solve
previously unsolved problems,
previously unsolved problems,
24
69925
3708
让人们尝试解决过去未解决的问题
01:25
and it's been successful
hundreds of times.
hundreds of times.
25
73633
3837
已成功解决问题几百次
01:29
So from this vantage point,
I was able to find out
I was able to find out
26
77470
2470
在这个有利环境中，我发现了
01:31
a lot about what machine learning
can do in the past, can do today,
can do in the past, can do today,
27
79940
3950
机器学习在过去，现在，和将来可以做些什么
01:35
and what it could do in the future.
28
83890
2362
01:38
Perhaps the first big success of
machine learning commercially was Google.
machine learning commercially was Google.
29
86252
4423
第一个机器学习的商业成功案例应该是谷歌
01:42
Google showed that it is
possible to find information
possible to find information
30
90675
3109
谷歌用计算机算法寻找信息
01:45
by using a computer algorithm,
31
93784
1752
01:47
and this algorithm is based
on machine learning.
on machine learning.
32
95536
2901
而且这个算法以计算机学习为基础
01:50
Since that time, there have been many
commercial successes of machine learning.
commercial successes of machine learning.
33
98437
3886
从那以后，机器学习得到了很多的商业成功
01:54
Companies like Amazon and Netflix
34
102323
1837
像亚马逊、网飞这类公司
01:56
use machine learning to suggest
products that you might like to buy,
products that you might like to buy,
35
104160
3716
通过机器学习向你推荐你可能想买的东西
01:59
movies that you might like to watch.
36
107876
2020
你可能想看的电影
02:01
Sometimes, it's almost creepy.
37
109896
1807
有时候你会被吓一跳
02:03
Companies like LinkedIn and Facebook
38
111703
1954
像领英、脸谱这类的公司
02:05
sometimes will tell you about
who your friends might be
who your friends might be
39
113657
2594
有时会告诉你谁会是你的朋友
02:08
and you have no idea how it did it,
40
116251
1977
你根本不知道他们是如何做到的
02:10
and this is because it's using
the power of machine learning.
the power of machine learning.
41
118228
2967
其实他们正是运用了机器学习的力量
02:13
These are algorithms that have
learned how to do this from data
learned how to do this from data
42
121195
2957
这种运算方法使用数据
02:16
rather than being programmed by hand.
43
124152
3247
而非手动编写程序
02:19
This is also how IBM was successful
44
127399
2478
这也是IBM的Watson超级计算机
在《危险边缘》里打败两届世界冠军的秘诀
在《危险边缘》里打败两届世界冠军的秘诀
02:21
in getting Watson to beat
the two world champions at "Jeopardy,"
the two world champions at "Jeopardy,"
45
129877
3862
02:25
answering incredibly subtle
and complex questions like this one.
and complex questions like this one.
46
133739
3225
成功回答了这样一个极其模糊且复杂的问题
02:28
["The ancient 'Lion of Nimrud' went missing
from this city's national museum in 2003
(along with a lot of other stuff)"]
from this city's national museum in 2003
(along with a lot of other stuff)"]
47
136964
2835
［“古代‘尼姆鲁德狮像’于2003年在这个城市的国家博物馆消失（连同其它很多物品）”］
02:31
This is also why we are now able
to see the first self-driving cars.
to see the first self-driving cars.
48
139799
3235
这也是为什么我们现在有了第一台自驾车
02:35
If you want to be able to tell
the difference between, say,
the difference between, say,
49
143034
2822
如果你想区分一棵树和一个行人
02:37
a tree and a pedestrian,
well, that's pretty important.
well, that's pretty important.
50
145856
2632
显然这很重要
02:40
We don't know how to write
those programs by hand,
those programs by hand,
51
148488
2587
但是我们不知道如何写这样一个程序
02:43
but with machine learning,
this is now possible.
this is now possible.
52
151075
2997
有了机器学习，这就成为了可能
02:46
And in fact, this car has driven
over a million miles
over a million miles
53
154072
2608
这台自驾车已经行驶了十万英里
02:48
without any accidents on regular roads.
54
156680
3506
在正常路面上零事故
02:52
So we now know that computers can learn,
55
160196
3914
我们知道电脑能够学习
02:56
and computers can learn to do things
56
164110
1900
学习做一件有时我们自己都不知道怎么做的事情
02:58
that we actually sometimes
don't know how to do ourselves,
don't know how to do ourselves,
57
166010
2838
03:00
or maybe can do them better than us.
58
168848
2885
有时甚至比我们做得更好
03:03
One of the most amazing examples
I've seen of machine learning
I've seen of machine learning
59
171733
4195
我见过机器学习最惊人的例子
是我在Kaggle做的一个项目
是我在Kaggle做的一个项目
03:07
happened on a project that I ran at Kaggle
60
175928
2392
03:10
where a team run by a guy
called Geoffrey Hinton
called Geoffrey Hinton
61
178320
3591
一个叫杰弗里 辛顿的人毕业于多伦多大学，
带领一个团队
带领一个团队
03:13
from the University of Toronto
62
181911
1552
03:15
won a competition for
automatic drug discovery.
automatic drug discovery.
63
183463
2677
赢得了一个自动查毒的竞赛
03:18
Now, what was extraordinary here
is not just that they beat
is not just that they beat
64
186140
2847
然而真正精彩的不是他们打败了所有默克公司
或者国际学术团体设计的运算
或者国际学术团体设计的运算
03:20
all of the algorithms developed by Merck
or the international academic community,
or the international academic community,
65
188987
4013
03:25
but nobody on the team had any background
in chemistry or biology or life sciences,
in chemistry or biology or life sciences,
66
193000
5061
而是他们团队里没有一个人有化学、生物
或者生命科学的背景
或者生命科学的背景
03:30
and they did it in two weeks.
67
198061
2169
却在两个星期内赢得了比赛
03:32
How did they do this?
68
200230
1381
他们是如何做到的？
03:34
They used an extraordinary algorithm
called deep learning.
called deep learning.
69
202421
2921
他们应用了一种超凡的算法叫做深度学习
03:37
So important was this that in fact
the success was covered
the success was covered
70
205342
2949
几个星期后纽约时报在其首页
报道了此次的重要成功
报道了此次的重要成功
03:40
in The New York Times in a front page
article a few weeks later.
article a few weeks later.
71
208291
3121
03:43
This is Geoffrey Hinton
here on the left-hand side.
here on the left-hand side.
72
211412
2735
在左手边就是杰弗里 辛顿
03:46
Deep learning is an algorithm
inspired by how the human brain works,
inspired by how the human brain works,
73
214147
4341
深度学习是受到人类大脑的启发
03:50
and as a result it's an algorithm
74
218488
1812
也因此这种算法的能力不受任何理论限制
03:52
which has no theoretical limitations
on what it can do.
on what it can do.
75
220300
3841
03:56
The more data you give it and the more
computation time you give it,
computation time you give it,
76
224141
2823
你给它越多的数据和运算时间
03:58
the better it gets.
77
226964
1312
它会工作的越好
04:00
The New York Times also
showed in this article
showed in this article
78
228276
2339
纽约时报在其文章中
还说明了深度学习的另一非凡之处
还说明了深度学习的另一非凡之处
04:02
another extraordinary
result of deep learning
result of deep learning
79
230615
2242
04:04
which I'm going to show you now.
80
232857
2712
现在我要展示给你们看
04:07
It shows that computers
can listen and understand.
can listen and understand.
81
235569
4941
它表明电脑能够听懂信息
04:12
(Video) Richard Rashid: Now, the last step
82
240510
2711
（视频）理查德 拉希德：现在，
我要做的最后一步是
我要做的最后一步是
04:15
that I want to be able
to take in this process
to take in this process
83
243221
3025
04:18
is to actually speak to you in Chinese.
84
246246
4715
用汉语和大家说话
04:22
Now the key thing there is,
85
250961
2635
在这之前，我们已经通过很多说汉语的人
收集了大量信息
收集了大量信息
04:25
we've been able to take a large amount
of information from many Chinese speakers
of information from many Chinese speakers
86
253596
5002
04:30
and produce a text-to-speech system
87
258598
2530
然后形成一个语音合成系统
04:33
that takes Chinese text
and converts it into Chinese language,
and converts it into Chinese language,
88
261128
4673
把汉字转换成汉语言
04:37
and then we've taken
an hour or so of my own voice
an hour or so of my own voice
89
265801
4128
之后我们收录了一个小时我的声音
04:41
and we've used that to modulate
90
269929
1891
使声音合成系统的声音听起来像我
04:43
the standard text-to-speech system
so that it would sound like me.
so that it would sound like me.
91
271820
4544
04:48
Again, the result's not perfect.
92
276364
2540
再次，结果并不完美
04:50
There are in fact quite a few errors.
93
278904
2648
他们会有不少错误
04:53
(In Chinese)
94
281552
2484
（中文）
04:56
(Applause)
95
284036
3367
（掌声）
05:01
There's much work to be done in this area.
96
289446
3576
在这个领域还有很多工作要做
05:05
(In Chinese)
97
293022
3645
（中文）
05:08
(Applause)
98
296667
3433
（掌声）
05:13
Jeremy Howard: Well, that was at
a machine learning conference in China.
a machine learning conference in China.
99
301345
3399
杰里米 霍华德：这是在一个中国的机器学习会议上
05:16
It's not often, actually,
at academic conferences
at academic conferences
100
304744
2370
事实上，一般来说，你不会在学术会议上
听到如此热烈的掌声
听到如此热烈的掌声
05:19
that you do hear spontaneous applause,
101
307114
1897
05:21
although of course sometimes
at TEDx conferences, feel free.
at TEDx conferences, feel free.
102
309011
3676
当然除了TEDx演讲可以随意鼓掌
05:24
Everything you saw there
was happening with deep learning.
was happening with deep learning.
103
312687
2795
你所看到的一切都伴随着深入学习
05:27
(Applause) Thank you.
104
315482
1525
（掌声）谢谢
05:29
The transcription in English
was deep learning.
was deep learning.
105
317007
2282
对英文的转录是深入学习
05:31
The translation to Chinese and the text
in the top right, deep learning,
in the top right, deep learning,
106
319289
3412
翻译成汉语以及屏幕右上方的文字是深入学习
05:34
and the construction of the voice
was deep learning as well.
was deep learning as well.
107
322701
3307
声音的合成也是深入学习
05:38
So deep learning is
this extraordinary thing.
this extraordinary thing.
108
326008
3234
深入学习就是这样神奇的事情
05:41
It's a single algorithm that
can seem to do almost anything,
can seem to do almost anything,
109
329242
3099
这个单一的算法似乎可以做任何事情
05:44
and I discovered that a year earlier,
it had also learned to see.
it had also learned to see.
110
332341
3111
而且一年前我发现他甚至有视觉
05:47
In this obscure competition from Germany
111
335452
2176
这个名不见经传的德国竞赛
05:49
called the German Traffic Sign
Recognition Benchmark,
Recognition Benchmark,
112
337628
2597
叫做德国交通标志识别基准
05:52
deep learning had learned
to recognize traffic signs like this one.
to recognize traffic signs like this one.
113
340225
3393
深度学习已学得识别这些交通标识
05:55
Not only could it
recognize the traffic signs
recognize the traffic signs
114
343618
2094
它不仅能够做的比其它算法好
05:57
better than any other algorithm,
115
345712
1758
05:59
the leaderboard actually showed
it was better than people,
it was better than people,
116
347470
2719
排行榜显示它比人更厉害
06:02
about twice as good as people.
117
350189
1852
是人的准确率的两倍
06:04
So by 2011, we had the first example
118
352041
1996
到2011年，我们有了第一台视力高于人类的电脑
06:06
of computers that can see
better than people.
better than people.
119
354037
3405
06:09
Since that time, a lot has happened.
120
357442
2049
从此更多的电脑也可以做到
06:11
In 2012, Google announced that
they had a deep learning algorithm
they had a deep learning algorithm
121
359491
3514
在2012年，谷歌宣布让一个深度学习的算法看YouTube视频
06:15
watch YouTube videos
122
363005
1415
06:16
and crunched the data
on 16,000 computers for a month,
on 16,000 computers for a month,
123
364420
3437
收集16，000台电脑上的数据，为期一个月
06:19
and the computer independently learned
about concepts such as people and cats
about concepts such as people and cats
124
367857
4361
之后电脑便能仅通过看视频独立识别人和猫
06:24
just by watching the videos.
125
372218
1809
06:26
This is much like the way
that humans learn.
that humans learn.
126
374027
2352
这近似于人类学习的过程
06:28
Humans don't learn
by being told what they see,
by being told what they see,
127
376379
2740
人类不需要被告诉他们看到了什么
06:31
but by learning for themselves
what these things are.
what these things are.
128
379119
3331
而是在自己认知事物的过程中学习
06:34
Also in 2012, Geoffrey Hinton,
who we saw earlier,
who we saw earlier,
129
382450
3369
同样在2012年，杰弗里 辛顿，我们之前看到的人
06:37
won the very popular ImageNet competition,
130
385819
2858
赢了很火的ImageNet比赛
06:40
looking to try to figure out
from one and a half million images
from one and a half million images
131
388677
4141
分辨出150万张图片的内容
06:44
what they're pictures of.
132
392818
1438
06:46
As of 2014, we're now down
to a six percent error rate
to a six percent error rate
133
394256
3533
到2014年，我们已经将图像识别的误差
降低到百分之六
降低到百分之六
06:49
in image recognition.
134
397789
1453
06:51
This is better than people, again.
135
399242
2026
低于人类误差率
06:53
So machines really are doing
an extraordinarily good job of this,
an extraordinarily good job of this,
136
401268
3769
这项非凡的工作现在已经用于工业
06:57
and it is now being used in industry.
137
405037
2269
06:59
For example, Google announced last year
138
407306
3042
比如说，去年谷歌声明
07:02
that they had mapped every single
location in France in two hours,
location in France in two hours,
139
410348
4585
他们在两小时内把法国的每一个地点汇成地图
07:06
and the way they did it was
that they fed street view images
that they fed street view images
140
414933
3447
他们是将街景填入深度学习算法以辨认街道号
07:10
into a deep learning algorithm
to recognize and read street numbers.
to recognize and read street numbers.
141
418380
4319
07:14
Imagine how long
it would have taken before:
it would have taken before:
142
422699
2220
可以想象从前这件事要花费多少时间和精力
07:16
dozens of people, many years.
143
424919
3355
07:20
This is also happening in China.
144
428274
1911
同样的事情也发生在中国
07:22
Baidu is kind of
the Chinese Google, I guess,
the Chinese Google, I guess,
145
430185
4036
百度大概类似于中国的谷歌
07:26
and what you see here in the top left
146
434221
2283
我们看到左上角
07:28
is an example of a picture that I uploaded
to Baidu's deep learning system,
to Baidu's deep learning system,
147
436504
3974
是一张我上传到百度的深度学习系统的图片
07:32
and underneath you can see that the system
has understood what that picture is
has understood what that picture is
148
440478
3769
下面你可以看到系统理解了这张照片
07:36
and found similar images.
149
444247
2236
并且找到了类似的图片
07:38
The similar images actually
have similar backgrounds,
have similar backgrounds,
150
446483
2736
同样的背景
07:41
similar directions of the faces,
151
449219
1658
同样的角度
07:42
even some with their tongue out.
152
450877
1788
有的甚至也有伸出来的舌头
07:44
This is not clearly looking
at the text of a web page.
at the text of a web page.
153
452665
3030
网页上没有准确的文字
07:47
All I uploaded was an image.
154
455695
1412
我只是上传了图片
07:49
So we now have computers which
really understand what they see
really understand what they see
155
457107
4021
所以说电脑能够真正理解它所看到的事物
07:53
and can therefore search databases
156
461128
1624
进而在数据库的几百万张图片中进行实时搜索
07:54
of hundreds of millions
of images in real time.
of images in real time.
157
462752
3554
07:58
So what does it mean
now that computers can see?
now that computers can see?
158
466306
3230
就现在而言，电脑的视力意味着什么呢？
08:01
Well, it's not just
that computers can see.
that computers can see.
159
469536
2017
事实上不仅仅是电脑能够看见
08:03
In fact, deep learning
has done more than that.
has done more than that.
160
471553
2069
深度学习其实可以做得更多
08:05
Complex, nuanced sentences like this one
161
473622
2948
像这样一个细小复杂的语句
08:08
are now understandable
with deep learning algorithms.
with deep learning algorithms.
162
476570
2824
对深度学习来说是相对易于理解的
08:11
As you can see here,
163
479394
1303
你可以看到
08:12
this Stanford-based system
showing the red dot at the top
showing the red dot at the top
164
480697
2768
斯坦福基础系统显示上面的红点指出
这个语句表达的是否定语气
这个语句表达的是否定语气
08:15
has figured out that this sentence
is expressing negative sentiment.
is expressing negative sentiment.
165
483465
3919
08:19
Deep learning now in fact
is near human performance
is near human performance
166
487384
3406
深度学习在理解语句内容方面已经接近人类水平
08:22
at understanding what sentences are about
and what it is saying about those things.
and what it is saying about those things.
167
490802
5121
08:27
Also, deep learning has
been used to read Chinese,
been used to read Chinese,
168
495923
2728
同样，深度学习在用于阅读汉语上已经相当于中国本土人水平
08:30
again at about native
Chinese speaker level.
Chinese speaker level.
169
498651
3156
08:33
This algorithm developed
out of Switzerland
out of Switzerland
170
501807
2168
这个算法开发于瑞士
08:35
by people, none of whom speak
or understand any Chinese.
or understand any Chinese.
171
503975
3356
没有一个人懂汉语
08:39
As I say, using deep learning
172
507331
2051
要我说，深度学习是比较于人类
做这件事最好的系统
做这件事最好的系统
08:41
is about the best system
in the world for this,
in the world for this,
173
509382
2219
08:43
even compared to native
human understanding.
human understanding.
174
511601
5117
08:48
This is a system that we
put together at my company
put together at my company
175
516718
2964
这个系统是在我们公司建立的
08:51
which shows putting
all this stuff together.
all this stuff together.
176
519682
2046
它要把这些东西集合起来
08:53
These are pictures which
have no text attached,
have no text attached,
177
521728
2461
这些图片没有文字描述
08:56
and as I'm typing in here sentences,
178
524189
2352
随着我在这输入文字
08:58
in real time it's understanding
these pictures
these pictures
179
526541
2969
同时它会了解这些图片
09:01
and figuring out what they're about
180
529510
1679
理解它们是关于什么的
09:03
and finding pictures that are similar
to the text that I'm writing.
to the text that I'm writing.
181
531189
3163
然后找出和这些相似的图片
09:06
So you can see, it's actually
understanding my sentences
understanding my sentences
182
534352
2756
所以你看，他真正在理解我的文字
09:09
and actually understanding these pictures.
183
537108
2224
理解这些图片
09:11
I know that you've seen
something like this on Google,
something like this on Google,
184
539332
2559
我知道你在谷歌上看到过类似的
09:13
where you can type in things
and it will show you pictures,
and it will show you pictures,
185
541891
2775
你可以输入文字，它会提供给你图片
09:16
but actually what it's doing is it's
searching the webpage for the text.
searching the webpage for the text.
186
544666
3424
但实际上它是在网页上搜索文字
09:20
This is very different from actually
understanding the images.
understanding the images.
187
548090
3001
这和理解图片是有很大不同的
09:23
This is something that computers
have only been able to do
have only been able to do
188
551091
2752
理解图片是电脑在过去几个月里才刚刚会做的事情
09:25
for the first time in the last few months.
189
553843
3248
09:29
So we can see now that computers
can not only see but they can also read,
can not only see but they can also read,
190
557091
4091
电脑不仅有视力，而且能够阅读
09:33
and, of course, we've shown that they
can understand what they hear.
can understand what they hear.
191
561182
3765
而且当然，电脑也能理解所听到的
09:36
Perhaps not surprising now that
I'm going to tell you they can write.
I'm going to tell you they can write.
192
564947
3442
也许并不意外，我现在要告诉你们，电脑也可以写
09:40
Here is some text that I generated
using a deep learning algorithm yesterday.
using a deep learning algorithm yesterday.
193
568389
4783
这是我昨天用深度学习算法写的文字
09:45
And here is some text that an algorithm
out of Stanford generated.
out of Stanford generated.
194
573172
3924
这些是斯坦福的算法做的
09:49
Each of these sentences was generated
195
577096
1764
每一句话都是深度学习算法对图片进行的描述
09:50
by a deep learning algorithm
to describe each of those pictures.
to describe each of those pictures.
196
578860
4249
09:55
This algorithm before has never seen
a man in a black shirt playing a guitar.
a man in a black shirt playing a guitar.
197
583109
4472
算法没见过一个穿黑衣服的男人弹吉他
09:59
It's seen a man before,
it's seen black before,
it's seen black before,
198
587581
2220
它见过男人，见过黑色
10:01
it's seen a guitar before,
199
589801
1599
见过吉他
10:03
but it has independently generated
this novel description of this picture.
this novel description of this picture.
200
591400
4294
它便自己对这个图片作出了这样的描述
10:07
We're still not quite at human
performance here, but we're close.
performance here, but we're close.
201
595694
3502
我们还做不到完全和人类同等水平，
但我们已经很接近了
但我们已经很接近了
10:11
In tests, humans prefer
the computer-generated caption
the computer-generated caption
202
599196
4068
统计表明，四分之一的人更喜欢电脑做的图片说明
10:15
one out of four times.
203
603264
1527
10:16
Now this system is now only two weeks old,
204
604791
2064
目前这个系统刚被开发两周之久
10:18
so probably within the next year,
205
606855
1846
所以按这个速度，估计明年
10:20
the computer algorithm will be
well past human performance
well past human performance
206
608701
2801
电脑算法会超过人类水平
10:23
at the rate things are going.
207
611502
1862
10:25
So computers can also write.
208
613364
3049
电脑会写
10:28
So we put all this together and it leads
to very exciting opportunities.
to very exciting opportunities.
209
616413
3475
我们把这些都放在一起，会发现一个令人兴奋的机遇
10:31
For example, in medicine,
210
619888
1492
比如说，在医药业
10:33
a team in Boston announced
that they had discovered
that they had discovered
211
621380
2525
一个波士顿团队宣布
10:35
dozens of new clinically relevant features
212
623905
2949
他们发现了肿瘤的几十种临床表现
10:38
of tumors which help doctors
make a prognosis of a cancer.
make a prognosis of a cancer.
213
626854
4266
帮助医生预测癌症
10:44
Very similarly, in Stanford,
214
632220
2296
同样的，在斯坦福
10:46
a group there announced that,
looking at tissues under magnification,
looking at tissues under magnification,
215
634516
3663
一个团队宣布通过用放大镜观察组织
10:50
they've developed
a machine learning-based system
a machine learning-based system
216
638179
2381
开发了一个基于机器学习的系统
10:52
which in fact is better
than human pathologists
than human pathologists
217
640560
2582
可以比病理学家更有效地预测癌症患者的幸存率
10:55
at predicting survival rates
for cancer sufferers.
for cancer sufferers.
218
643142
4377
10:59
In both of these cases, not only
were the predictions more accurate,
were the predictions more accurate,
219
647519
3245
在这两个例子中，不仅预测更加准确
11:02
but they generated new insightful science.
220
650764
2502
而且他们创造了新的科学视角
11:05
In the radiology case,
221
653276
1505
在放射学中
11:06
they were new clinical indicators
that humans can understand.
that humans can understand.
222
654781
3095
新视角是人类可以明白的新临床表现
11:09
In this pathology case,
223
657876
1792
在病理学中
11:11
the computer system actually discovered
that the cells around the cancer
that the cells around the cancer
224
659668
4500
电脑发现癌细胞周围的细胞
11:16
are as important as
the cancer cells themselves
the cancer cells themselves
225
664168
3340
在诊断中同癌细胞一样重要
11:19
in making a diagnosis.
226
667508
1752
11:21
This is the opposite of what pathologists
had been taught for decades.
had been taught for decades.
227
669260
5361
这和病理学家几十年来的教学是相反的
11:26
In each of those two cases,
they were systems developed
they were systems developed
228
674621
3292
这两个案例中的系统都是由
11:29
by a combination of medical experts
and machine learning experts,
and machine learning experts,
229
677913
3621
医学专家和机器学习专家共同开发的
11:33
but as of last year,
we're now beyond that too.
we're now beyond that too.
230
681534
2741
去年我们就已经超过了这个水平
11:36
This is an example of
identifying cancerous areas
identifying cancerous areas
231
684275
3549
这个是用显微镜识别组织癌变区的例子
11:39
of human tissue under a microscope.
232
687824
2530
11:42
The system being shown here
can identify those areas more accurately,
can identify those areas more accurately,
233
690354
4613
所显示的这个系统能够与病理学专家同样准确地识别癌变区
11:46
or about as accurately,
as human pathologists,
as human pathologists,
234
694967
2775
甚至比病理专家更准确
11:49
but was built entirely with deep learning
using no medical expertise
using no medical expertise
235
697742
3392
但是建立系统的都是深度学习的专家
11:53
by people who have
no background in the field.
no background in the field.
236
701134
2526
没有一个医学专家
11:56
Similarly, here, this neuron segmentation.
237
704730
2555
类似的，这是神经细胞分裂
11:59
We can now segment neurons
about as accurately as humans can,
about as accurately as humans can,
238
707285
3668
我们已经可以和人类一样准确地分裂细胞
12:02
but this system was developed
with deep learning
with deep learning
239
710953
2717
但这是个深度学习系统
12:05
using people with no previous
background in medicine.
background in medicine.
240
713670
3251
没有一个开发者拥有医学背景
12:08
So myself, as somebody with
no previous background in medicine,
no previous background in medicine,
241
716921
3227
对于我这个完全没有医学背景的人来说
12:12
I seem to be entirely well qualified
to start a new medical company,
to start a new medical company,
242
720148
3727
看起来我也完全可以开一个医药公司
12:15
which I did.
243
723875
2146
我确实这么做了
12:18
I was kind of terrified of doing it,
244
726021
1740
我开始有点不知所措
12:19
but the theory seemed to suggest
that it ought to be possible
that it ought to be possible
245
727761
2889
但理论上说这件事是可行的
12:22
to do very useful medicine
using just these data analytic techniques.
using just these data analytic techniques.
246
730650
5492
用这些数据分析技术制作医药
12:28
And thankfully, the feedback
has been fantastic,
has been fantastic,
247
736142
2480
所幸的是，反响非常好
12:30
not just from the media
but from the medical community,
but from the medical community,
248
738622
2356
不仅是媒体的，包括医药行业
12:32
who have been very supportive.
249
740978
2344
都很支持
12:35
The theory is that we can take
the middle part of the medical process
the middle part of the medical process
250
743322
4149
理论表明我们可以将制药的中间过程
12:39
and turn that into data analysis
as much as possible,
as much as possible,
251
747471
2893
充分转换成数据分析
12:42
leaving doctors to do
what they're best at.
what they're best at.
252
750364
3065
让医生去做他们最擅长的
12:45
I want to give you an example.
253
753429
1602
我有一个例子
12:47
It now takes us about 15 minutes
to generate a new medical diagnostic test
to generate a new medical diagnostic test
254
755031
4944
制作一个医学诊断测试需要十五分钟
12:51
and I'll show you that in real time now,
255
759975
1954
我会给你们实际展示
12:53
but I've compressed it down to
three minutes by cutting some pieces out.
three minutes by cutting some pieces out.
256
761929
3487
但是我去掉了一部分，把它压缩到了三分钟
12:57
Rather than showing you
creating a medical diagnostic test,
creating a medical diagnostic test,
257
765416
3061
不要医学诊断试验
13:00
I'm going to show you
a diagnostic test of car images,
a diagnostic test of car images,
258
768477
3369
我要给你们展示制作一个汽车图片的诊断测试
13:03
because that's something
we can all understand.
we can all understand.
259
771846
2222
因为这个我们都能懂
13:06
So here we're starting with
about 1.5 million car images,
about 1.5 million car images,
260
774068
3201
现在我们有150万张汽车图片
13:09
and I want to create something
that can split them into the angle
that can split them into the angle
261
777269
3206
我想要根据拍照的角度对他们进行分类
13:12
of the photo that's being taken.
262
780475
2223
13:14
So these images are entirely unlabeled,
so I have to start from scratch.
so I have to start from scratch.
263
782698
3888
这些图片完全没有标签，所以我要先对他们进行简单描述
13:18
With our deep learning algorithm,
264
786586
1865
有深度学习算法
13:20
it can automatically identify
areas of structure in these images.
areas of structure in these images.
265
788451
3707
它可以自动识别图片的结构要素
13:24
So the nice thing is that the human
and the computer can now work together.
and the computer can now work together.
266
792158
3620
令人高兴的是人和电脑可以合作
13:27
So the human, as you can see here,
267
795778
2178
你可以看到，这个人
13:29
is telling the computer
about areas of interest
about areas of interest
268
797956
2675
正在告诉电脑什么是感兴趣的要素
13:32
which it wants the computer then
to try and use to improve its algorithm.
to try and use to improve its algorithm.
269
800631
4650
为之后电脑用来完善算法
13:37
Now, these deep learning systems actually
are in 16,000-dimensional space,
are in 16,000-dimensional space,
270
805281
4296
现在，这些深度学习算法处在16，000维空间中
13:41
so you can see here the computer
rotating this through that space,
rotating this through that space,
271
809577
3432
所以你看到电脑让他们在这个空间中旋转
13:45
trying to find new areas of structure.
272
813009
1992
尝试找到新的结构要素
13:47
And when it does so successfully,
273
815001
1781
当他成功时
13:48
the human who is driving it can then
point out the areas that are interesting.
point out the areas that are interesting.
274
816782
4004
开车的人就可以指出感兴趣的要素
13:52
So here, the computer has
successfully found areas,
successfully found areas,
275
820786
2422
现在电脑成功找出这些要素
13:55
for example, angles.
276
823208
2562
比如，角度
13:57
So as we go through this process,
277
825770
1606
我们在这个过程中
13:59
we're gradually telling
the computer more and more
the computer more and more
278
827376
2340
逐渐的告诉电脑更多
14:01
about the kinds of structures
we're looking for.
we're looking for.
279
829716
2428
我们想寻找的结构
14:04
You can imagine in a diagnostic test
280
832144
1772
你可以想象一个诊断测试
14:05
this would be a pathologist identifying
areas of pathosis, for example,
areas of pathosis, for example,
281
833916
3350
这就像是病理学家识别病态区域
14:09
or a radiologist indicating
potentially troublesome nodules.
potentially troublesome nodules.
282
837266
5026
或者放射学专家找出潜在的问题囊肿
14:14
And sometimes it can be
difficult for the algorithm.
difficult for the algorithm.
283
842292
2559
有时候这对算法来说有些难度
14:16
In this case, it got kind of confused.
284
844851
1964
我们的例子就比较麻烦
14:18
The fronts and the backs
of the cars are all mixed up.
of the cars are all mixed up.
285
846815
2550
车的正面和背面全部混淆了
14:21
So here we have to be a bit more careful,
286
849365
2072
所以我们要仔细一些
14:23
manually selecting these fronts
as opposed to the backs,
as opposed to the backs,
287
851437
3232
人工地选出正面和背面
14:26
then telling the computer
that this is a type of group
that this is a type of group
288
854669
5506
人后告诉电脑这是我们所感兴趣的一类
14:32
that we're interested in.
289
860175
1348
14:33
So we do that for a while,
we skip over a little bit,
we skip over a little bit,
290
861523
2677
做这件事花了一些时间，所以我们跳过
14:36
and then we train the
machine learning algorithm
machine learning algorithm
291
864200
2246
之后我们用这几百个东西训练机器学习算法
14:38
based on these couple of hundred things,
292
866446
1974
14:40
and we hope that it's gotten a lot better.
293
868420
2025
希望他会有很大进步
14:42
You can see, it's now started to fade
some of these pictures out,
some of these pictures out,
294
870445
3073
你能看到，它正在消退一些图片
14:45
showing us that it already is recognizing
how to understand some of these itself.
how to understand some of these itself.
295
873518
4708
说明他已经开始可以自己理解这些图片了
14:50
We can then use this concept
of similar images,
of similar images,
296
878226
2902
我们可以用相似图片的概念
14:53
and using similar images, you can now see,
297
881128
2094
用相似的图片，你可以看到
14:55
the computer at this point is able to
entirely find just the fronts of cars.
entirely find just the fronts of cars.
298
883222
4019
电脑现在能够只找出正面的车
14:59
So at this point, the human
can tell the computer,
can tell the computer,
299
887241
2948
在这个时候，人可以告诉电脑
15:02
okay, yes, you've done
a good job of that.
a good job of that.
300
890189
2293
对的，没错，你做的很好
15:05
Sometimes, of course, even at this point
301
893652
2185
当然，有时，即使在这个阶段
15:07
it's still difficult
to separate out groups.
to separate out groups.
302
895837
3674
分组仍然是很困难的
15:11
In this case, even after we let the
computer try to rotate this for a while,
computer try to rotate this for a while,
303
899511
3884
像我们这里，让电脑在这里旋转了一段时间了
15:15
we still find that the left sides
and the right sides pictures
and the right sides pictures
304
903399
3345
我们还是看到左面的和右面的图片有混淆
15:18
are all mixed up together.
305
906744
1478
15:20
So we can again give
the computer some hints,
the computer some hints,
306
908222
2140
所以我们可以再一次给电脑一些提示
15:22
and we say, okay, try and find
a projection that separates out
a projection that separates out
307
910362
2976
我们让它通过深度学习算法尽可能分离出左面和右面的图片
15:25
the left sides and the right sides
as much as possible
as much as possible
308
913338
2607
15:27
using this deep learning algorithm.
309
915945
2122
15:30
And giving it that hint --
ah, okay, it's been successful.
ah, okay, it's been successful.
310
918067
2942
有了这个指示——好的，它已经完成了
15:33
It's managed to find a way
of thinking about these objects
of thinking about these objects
311
921009
2882
它要想办法分开这一部分
15:35
that's separated out these together.
312
923891
2380
15:38
So you get the idea here.
313
926271
2438
你现在知道了
15:40
This is a case not where the human
is being replaced by a computer,
is being replaced by a computer,
314
928709
8197
这不是电脑取代人类
15:48
but where they're working together.
315
936906
2640
而是一起合作
15:51
What we're doing here is we're replacing
something that used to take a team
something that used to take a team
316
939546
3550
我们在做的是将过去需要五六人的团队
用七年时间做的事情
用七年时间做的事情
15:55
of five or six people about seven years
317
943096
2002
15:57
and replacing it with something
that takes 15 minutes
that takes 15 minutes
318
945098
2605
变成只需一个人花十五分钟就能完成
15:59
for one person acting alone.
319
947703
2505
16:02
So this process takes about
four or five iterations.
four or five iterations.
320
950208
3950
这个过程需要四到五次反复
16:06
You can see we now have 62 percent
321
954158
1859
你可以看到我们已经将150万张图片的62%正确分类
16:08
of our 1.5 million images
classified correctly.
classified correctly.
322
956017
2959
16:10
And at this point, we
can start to quite quickly
can start to quite quickly
323
958976
2472
现在我们就可以快速地检查整个分组
16:13
grab whole big sections,
324
961448
1297
16:14
check through them to make sure
that there's no mistakes.
that there's no mistakes.
325
962745
2919
确保没有错误
16:17
Where there are mistakes, we can
let the computer know about them.
let the computer know about them.
326
965664
3952
如果哪里有错误，我们可以告诉电脑
16:21
And using this kind of process
for each of the different groups,
for each of the different groups,
327
969616
3045
每个分组我们都这样做
16:24
we are now up to
an 80 percent success rate
an 80 percent success rate
328
972661
2487
现在这150万张图片已经达到80%的成功率
16:27
in classifying the 1.5 million images.
329
975148
2415
16:29
And at this point, it's just a case
330
977563
2078
现在这个阶段
16:31
of finding the small number
that aren't classified correctly,
that aren't classified correctly,
331
979641
3579
只需要找出几个不正确的分类
16:35
and trying to understand why.
332
983220
2888
并让电脑明白为什么
16:38
And using that approach,
333
986108
1743
到了这个步骤
16:39
by 15 minutes we get
to 97 percent classification rates.
to 97 percent classification rates.
334
987851
4121
十五分钟后我们达到了97%的正确率
16:43
So this kind of technique
could allow us to fix a major problem,
could allow us to fix a major problem,
335
991972
4600
这种技术能帮助我们解决一个问题
16:48
which is that there's a lack
of medical expertise in the world.
of medical expertise in the world.
336
996578
3036
医疗专家不足的问题
16:51
The World Economic Forum says
that there's between a 10x and a 20x
that there's between a 10x and a 20x
337
999614
3489
世界经济论坛表明，在发展中国家，
内科医生有十倍到二十倍的短缺
内科医生有十倍到二十倍的短缺
16:55
shortage of physicians
in the developing world,
in the developing world,
338
1003103
2624
16:57
and it would take about 300 years
339
1005727
2113
而弥补这一短缺需要300年的时间
16:59
to train enough people
to fix that problem.
to fix that problem.
340
1007840
2894
17:02
So imagine if we can help
enhance their efficiency
enhance their efficiency
341
1010734
2885
所以想象一下，是否我们能够用深度学习的方法
帮助他们提高效率？
帮助他们提高效率？
17:05
using these deep learning approaches?
342
1013619
2839
17:08
So I'm very excited
about the opportunities.
about the opportunities.
343
1016458
2232
我对这个机会表示很激动
17:10
I'm also concerned about the problems.
344
1018690
2589
我同样的担心一些问题
17:13
The problem here is that
every area in blue on this map
every area in blue on this map
345
1021279
3124
问题是在这张地图上的蓝色区域内
17:16
is somewhere where services
are over 80 percent of employment.
are over 80 percent of employment.
346
1024403
3769
服务占就业的80%以上
17:20
What are services?
347
1028172
1787
什么是服务？
17:21
These are services.
348
1029959
1514
这些是服务
17:23
These are also the exact things that
computers have just learned how to do.
computers have just learned how to do.
349
1031473
4154
这些也是电脑才刚刚开始学习的事情
17:27
So 80 percent of the world's employment
in the developed world
in the developed world
350
1035627
3804
也就是说世界上发达国家的80%的就业
17:31
is stuff that computers
have just learned how to do.
have just learned how to do.
351
1039431
2532
是电脑刚开始学习的
17:33
What does that mean?
352
1041963
1440
这是什么意思？
17:35
Well, it'll be fine.
They'll be replaced by other jobs.
They'll be replaced by other jobs.
353
1043403
2583
其实也没什么大不了的，他们会被其他职业替代
17:37
For example, there will be
more jobs for data scientists.
more jobs for data scientists.
354
1045986
2707
比如说会有更多的数据学家
17:40
Well, not really.
355
1048693
817
也不尽然
17:41
It doesn't take data scientists
very long to build these things.
very long to build these things.
356
1049510
3118
数据学家不需要太久的时间做这些事
17:44
For example, these four algorithms
were all built by the same guy.
were all built by the same guy.
357
1052628
3252
比如这四个算法都是同时一个人开发的
17:47
So if you think, oh,
it's all happened before,
it's all happened before,
358
1055880
2438
如果你认为这些曾经都发生过
17:50
we've seen the results in the past
of when new things come along
of when new things come along
359
1058318
3808
我们看到过新的事物出现
17:54
and they get replaced by new jobs,
360
1062126
2252
然后被新的职业所取代
17:56
what are these new jobs going to be?
361
1064378
2116
那这些新的职业又会是什么？
17:58
It's very hard for us to estimate this,
362
1066494
1871
很难做出估计
18:00
because human performance
grows at this gradual rate,
grows at this gradual rate,
363
1068365
2739
因为人的能力以这个均匀的速度增长
18:03
but we now have a system, deep learning,
364
1071104
2562
但是现在我们有了深度学习系统
18:05
that we know actually grows
in capability exponentially.
in capability exponentially.
365
1073666
3227
它的能力以指数方式增长
18:08
And we're here.
366
1076893
1605
我们现在在这
18:10
So currently, we see the things around us
367
1078498
2061
目前，我们看周围的事物
18:12
and we say, "Oh, computers
are still pretty dumb." Right?
are still pretty dumb." Right?
368
1080559
2676
会说：“电脑还是很笨。”对吧？
18:15
But in five years' time,
computers will be off this chart.
computers will be off this chart.
369
1083235
3429
但是在五年内，电脑会超出这张图
18:18
So we need to be starting to think
about this capability right now.
about this capability right now.
370
1086664
3865
所以我们现在要开始考虑这样的能力了
18:22
We have seen this once before, of course.
371
1090529
2050
当然，我们曾经见过这个
18:24
In the Industrial Revolution,
372
1092579
1387
在工业革命时期
18:25
we saw a step change
in capability thanks to engines.
in capability thanks to engines.
373
1093966
2851
发动机让生产力迈进一大步
18:29
The thing is, though,
that after a while, things flattened out.
that after a while, things flattened out.
374
1097667
3138
然而问题是，一段时间之后，形势转平了
18:32
There was social disruption,
375
1100805
1702
是由于社会的破坏
18:34
but once engines were used
to generate power in all the situations,
to generate power in all the situations,
376
1102507
3439
但当发动机被普遍应用时
18:37
things really settled down.
377
1105946
2354
一切都稳定下来了
18:40
The Machine Learning Revolution
378
1108300
1473
机器学习革命
18:41
is going to be very different
from the Industrial Revolution,
from the Industrial Revolution,
379
1109773
2909
将和工业革命有很大不同
18:44
because the Machine Learning Revolution,
it never settles down.
it never settles down.
380
1112682
2950
因为机器学习革命不会停止
18:47
The better computers get
at intellectual activities,
at intellectual activities,
381
1115632
2982
电脑越擅长智能活动
18:50
the more they can build better computers
to be better at intellectual capabilities,
to be better at intellectual capabilities,
382
1118614
4248
它们越能制造出更加擅长智能活动的电脑
18:54
so this is going to be a kind of change
383
1122862
1908
这将会是世界从未经历过的改变
18:56
that the world has actually
never experienced before,
never experienced before,
384
1124770
2478
18:59
so your previous understanding
of what's possible is different.
of what's possible is different.
385
1127248
3306
所以你之前理解的可能性是不一样的
19:02
This is already impacting us.
386
1130974
1780
这正在影响我们的生活
19:04
In the last 25 years,
as capital productivity has increased,
as capital productivity has increased,
387
1132754
3630
在过去的25年里，随着资本生产力的增加
19:08
labor productivity has been flat,
in fact even a little bit down.
in fact even a little bit down.
388
1136400
4188
劳动生产力在变缓，甚至下降
19:13
So I want us to start
having this discussion now.
having this discussion now.
389
1141408
2741
所以我希望可以发起大家的讨论
19:16
I know that when I often tell people
about this situation,
about this situation,
390
1144149
3027
我知道当我和人们讲述这样的处境时
19:19
people can be quite dismissive.
391
1147176
1490
人们往往表现出不以为然
19:20
Well, computers can't really think,
392
1148666
1673
电脑不会思考
19:22
they don't emote,
they don't understand poetry,
they don't understand poetry,
393
1150339
3028
它们没有情感，也不懂诗
19:25
we don't really understand how they work.
394
1153367
2521
它们甚至都不知道自己是如何运作的
19:27
So what?
395
1155888
1486
那又怎样？
19:29
Computers right now can do the things
396
1157374
1804
电脑现在可以做
19:31
that humans spend most
of their time being paid to do,
of their time being paid to do,
397
1159178
2719
人类用大部分有偿的劳动时间做的事情
19:33
so now's the time to start thinking
398
1161897
1731
所以现在该到我们思考
19:35
about how we're going to adjust our
social structures and economic structures
social structures and economic structures
399
1163628
4387
我们将如何调整我们的社会结构和经济结构
19:40
to be aware of this new reality.
400
1168015
1840
来应对新形势
19:41
Thank you.
401
1169855
1533
谢谢
19:43
(Applause)
402
1171388
802
（鼓掌）
ABOUT THE SPEAKER
Jeremy Howard - Data scientistJeremy Howard imagines how advanced machine learning can improve our lives.
Why you should listen
Jeremy Howard is the CEO of Enlitic, an advanced machine learning company in San Francisco. Previously, he was the president and chief scientist at Kaggle, a community and competition platform of over 200,000 data scientists. Howard is a faculty member at Singularity University, where he teaches data science. He is also a Young Global Leader with the World Economic Forum, and spoke at the World Economic Forum Annual Meeting 2014 on "Jobs for the Machines."
Howard advised Khosla Ventures as their Data Strategist, identifying opportunities for investing in data-driven startups and mentoring portfolio companies to build data-driven businesses. He was the founding CEO of two successful Australian startups, FastMail and Optimal Decisions Group.
Jeremy Howard | Speaker | TED.com