ABOUT THE SPEAKER
Vittorio Loreto - Physicist
Vittorio Loreto is passionate about the complexity of the world around us in all its forms and he actively tries to decode it.

Why you should listen

Vittorio Loreto is a physicist at Sapienza University of Rome and faculty of the Complexity Science Hub Vienna. He is presently director of the SONY Computer Science Laboratories in Paris where he heads the team on creativity, innovation and artificial intelligence. He recently coordinated the research program dubbed KREYON, aimed at unfolding the dynamics of creativity, novelties and innovation. While theoretical modeling and data analysis are his native research tools, in the last few years he has been developing interactive tools, games, installations, to directly involve the public on the very research agenda. He created the KREYON DAYS, a new form of scientific event that tightly entangles research, learning, awareness and fun.

More profile about the speaker
Vittorio Loreto | Speaker | TED.com
TED@BCG Milan

Vittorio Loreto: Need a new idea? Start at the edge of what is known

Filmed:
1,475,309 views

"Where do great ideas come from?" Starting with this question in mind, Vittorio Loreto takes us on a journey to explore a possible mathematical scheme that explains the birth of the new. Learn more about the "adjacent possible" -- the crossroads of what's actual and what's possible -- and how studying the math that drives it could explain how we create new ideas.
- Physicist
Vittorio Loreto is passionate about the complexity of the world around us in all its forms and he actively tries to decode it. Full bio

Double-click the English transcript below to play the video.

00:14
We have all probably wondered
0
2349
2867
00:17
how great minds achieved
what they achieved, right?
1
5240
4176
00:21
And the more astonishing
their achievements are,
2
9440
2656
00:24
the more we call them geniuses,
3
12120
2536
00:26
perhaps aliens
4
14680
1536
00:28
coming from a different planet,
5
16239
2097
00:30
definitely not someone like us.
6
18360
2656
00:33
But is that true?
7
21040
1776
00:34
So let me start with an example.
8
22840
1800
00:37
You all know the story
of Newton's apple, right? OK.
9
25440
3816
00:41
Is that true? Probably not.
10
29280
2936
00:44
Still, it's difficult to think
that no apple at all was there.
11
32240
5216
00:49
I mean some stepping stone,
some specific conditions
12
37480
3616
00:53
that made universal gravitation
not impossible to conceive.
13
41120
4016
00:57
And definitely this was not impossible,
14
45160
2376
00:59
at least for Newton.
15
47560
1576
01:01
It was possible,
16
49160
1256
01:02
and for some reason, it was also there,
17
50440
3056
01:05
available at some point,
easy to pick as an apple.
18
53520
3776
01:09
Here is the apple.
19
57320
1616
01:10
And what about Einstein?
20
58960
2216
01:13
Was relativity theory another big leap
in the history of ideas
21
61200
5296
01:18
no one else could even conceive?
22
66520
2656
01:21
Or rather, was it again
something adjacent and possible,
23
69200
4456
01:25
to Einstein of course,
24
73680
2096
01:27
and he got there by small steps
and his very peculiar scientific path?
25
75800
4216
01:32
Of course we cannot conceive this path,
26
80040
2456
01:34
but this doesn't mean
that the path was not there.
27
82520
2480
01:38
So all of this seems very evocative,
28
86760
4856
01:43
but I would say hardly concrete
29
91640
1536
01:45
if we really want to grasp
the origin of great ideas
30
93200
3576
01:48
and more generally the way
in which the new enters our lives.
31
96800
4016
01:52
As a physicist, as a scientist,
32
100840
1976
01:54
I have learned that posing
the right questions
33
102840
2176
01:57
is half of the solution.
34
105040
2016
01:59
But I think now we start having
a great conceptual framework
35
107080
4736
02:03
to conceive and address
the right questions.
36
111840
3176
02:07
So let me drive you
to the edge of what is known,
37
115040
3456
02:10
or at least, what I know,
38
118520
2096
02:12
and let me show you that what is known
39
120640
2056
02:14
could be a powerful
and fascinating starting point
40
122720
4576
02:19
to grasp the deep meaning
of words like novelty, innovation,
41
127320
5096
02:24
creativity perhaps.
42
132440
1560
02:26
So we are discussing the "new,"
43
134880
3336
02:30
and of course, the science behind it.
44
138240
2656
02:32
The new can enter our lives
in many different ways,
45
140920
2976
02:35
can be very personal,
46
143920
1696
02:37
like I meet a new person,
47
145640
1936
02:39
I read a new book,
or I listen to a new song.
48
147600
3296
02:42
Or it could be global,
49
150920
1256
02:44
I mean, something we call innovation.
50
152200
2056
02:46
It could be a new theory,
a new technology,
51
154280
2176
02:48
but it could also be a new book
if you're the writer,
52
156480
2576
02:51
or it could be a new song
if you're the composer.
53
159080
2336
02:53
In all of these global cases,
the new is for everyone,
54
161440
4296
02:57
but experiencing the new
can be also frightening,
55
165760
3816
03:01
so the new can also frighten us.
56
169600
3736
03:05
But still, experiencing the new
means exploring a very peculiar space,
57
173360
4176
03:09
the space of what could be,
58
177560
2096
03:11
the space of the possible,
the space of possibilities.
59
179680
3176
03:14
It's a very weird space,
so I'll try to get you through this space.
60
182880
3456
03:18
So it could be a physical space.
61
186360
2016
03:20
So in this case, for instance,
62
188400
1616
03:22
novelty could be climbing
Machu Picchu for the first time,
63
190040
4056
03:26
as I did in 2016.
64
194120
1920
03:28
It could be a conceptual space,
65
196960
1816
03:30
so acquiring new information,
making sense of it, in a word, learning.
66
198800
4416
03:35
It could be a biological space.
67
203240
1936
03:37
I mean, think about the never-ending
fight of viruses and bacteria
68
205200
4096
03:41
with our immune system.
69
209320
1936
03:43
And now comes the bad news.
70
211280
1736
03:45
We are very, very bad
at grasping this space.
71
213040
3296
03:48
Think of it. Let's make an experiment.
72
216360
2016
03:50
Try to think about all the possible things
you could do in the next, say, 24 hours.
73
218400
6880
03:58
Here the key word is "all."
74
226320
2656
04:01
Of course you can conceive a few options,
like having a drink, writing a letter,
75
229000
4800
04:06
also sleeping during this boring talk,
76
234840
3176
04:10
if you can.
77
238040
1696
04:11
But not all of them.
78
239760
1656
04:13
So think about an alien invasion,
now, here, in Milan,
79
241440
3936
04:17
or me -- I stopped thinking
for 15 minutes.
80
245400
3120
04:21
So it's very difficult
to conceive this space,
81
249440
3136
04:24
but actually we have an excuse.
82
252600
2176
04:26
So it's not so easy to conceive this space
83
254800
3496
04:30
because we are trying to conceive
the occurrence of something brand new,
84
258320
3495
04:33
so something that never occurred before,
85
261839
1977
04:35
so we don't have clues.
86
263840
1480
04:38
A typical solution could be
87
266040
2896
04:40
looking at the future
with the eyes of the past,
88
268960
3216
04:44
so relying on all
the time series of past events
89
272200
3296
04:47
and hoping that this is enough
to predict the future.
90
275520
3496
04:51
But we know this is not working.
91
279040
2176
04:53
For instance, this was the first attempt
for weather forecasts, and it failed.
92
281240
5216
04:58
And it failed because
of the great complexity
93
286480
2416
05:00
of the underlying phenomenon.
94
288920
1936
05:02
So now we know that predictions
had to be based on modeling,
95
290880
5616
05:08
which means creating
a synthetic model of the system,
96
296520
3496
05:12
simulating this model
and then projecting the system
97
300040
4136
05:16
into the future through this model.
98
304200
2536
05:18
And now we can do this in a lot of cases
99
306760
2936
05:21
with the help of a lot of data.
100
309720
1880
05:25
Looking at the future
with the eye of the past
101
313000
2896
05:27
could be misleading also for machines.
102
315920
2736
05:30
Think about it.
103
318680
1216
05:31
Now picture yourself for a second
in the middle of the Australian Outback.
104
319920
4800
05:37
You stand there under the sun.
105
325440
2720
05:40
So you see something weird happening.
106
328840
2216
05:43
The car suddenly stops
107
331080
2736
05:45
very, very far from a kangaroo
crossing the street.
108
333840
3056
05:48
You look closer
109
336920
1456
05:50
and you realize
that the car has no driver.
110
338400
2416
05:52
It is not restarting, even after
the kangaroo is not there anymore.
111
340840
4016
05:56
So for some reasons,
112
344880
1896
05:58
the algorithms driving the car
cannot make sense
113
346800
2536
06:01
of this strange beast
jumping here and there on the street.
114
349360
3680
06:05
So it just stops.
115
353640
1200
06:07
Now, I should tell you,
this is a true story.
116
355720
2136
06:09
It happened a few months ago
to Volvo's self-driving cars
117
357880
2696
06:12
in the middle of the Australian Outback.
118
360600
1936
06:14
(Laughter)
119
362560
1696
06:16
It is a general problem,
120
364280
1976
06:18
and I guess this will affect
more and more in the near future
121
366280
2976
06:21
artificial intelligence
and machine learning.
122
369280
2560
06:24
It's also a very old problem,
I would say 17th century,
123
372440
3976
06:28
but I guess now we have new tools
and new clues to start solving it.
124
376440
5136
06:33
So let me take a step back,
125
381600
2176
06:35
five years back.
126
383800
2736
06:38
Italy. Rome. Winter.
127
386560
2976
06:41
So the winter of 2012
was very special in Rome.
128
389560
3576
06:45
Rome witnessed one of the greatest
snowfalls of its history.
129
393160
3560
06:49
That winter was special also
for me and my colleagues,
130
397520
3696
06:53
because we had an insight
about the possible mathematical scheme --
131
401240
3496
06:56
again, possible,
possible mathematical scheme,
132
404760
2976
06:59
to conceive the occurrence of the new.
133
407760
1840
07:02
I remember that day
because it was snowing,
134
410520
2416
07:04
so due to the snowfall,
we were blocked, stuck in my department,
135
412960
3776
07:08
and we couldn't go home,
136
416760
1416
07:10
so we got another coffee, we relaxed
137
418200
3056
07:13
and we kept discussing.
138
421280
1776
07:15
But at some point --
maybe not that date, precisely --
139
423080
3696
07:18
at some point we made the connection
140
426800
2896
07:21
between the problem of the new
141
429720
2976
07:24
and a beautiful concept
proposed years before
142
432720
2416
07:27
by Stuart Kauffman,
143
435160
1776
07:28
the adjacent possible.
144
436960
2040
07:31
So the adjacent possible
consists of all those things.
145
439720
3056
07:34
It could be ideas, it could be molecules,
it could be technological products
146
442800
3736
07:38
that are one step away
147
446560
2936
07:41
from what actually exists,
148
449520
1736
07:43
and you can achieve them
through incremental modifications
149
451280
3536
07:46
and recombinations
of the existing material.
150
454840
2560
07:50
So for instance, if I speak
about the space of my friends,
151
458520
3896
07:54
my adjacent possible would be
the set of all friends of my friends
152
462440
3976
07:58
not already my friends.
153
466440
1400
08:00
I hope that's clear.
154
468240
1736
08:02
But now if I meet a new person,
155
470000
1816
08:03
say Briar,
156
471840
1696
08:05
all her friends would immediately enter
my adjacent possible,
157
473560
4056
08:09
pushing its boundaries further.
158
477640
1520
08:12
So if you really want to look
from the mathematical point of view --
159
480160
3216
08:15
I'm sure you want --
160
483400
1400
08:18
you can actually look at this picture.
161
486200
1976
08:20
So suppose now this is your universe.
162
488200
1896
08:22
I know I'm asking a lot.
163
490120
1256
08:23
I mean, this is your universe.
Now you are the red spot.
164
491400
2640
08:27
And the green spot
is the adjacent possible for you,
165
495320
2616
08:29
so something you've never touched before.
166
497960
2096
08:32
So you do your normal life.
167
500080
1336
08:33
You move. You move in the space.
168
501440
1656
08:35
You have a drink.
You meet friends. You read a book.
169
503120
2656
08:37
At some point,
you end up on the green spot,
170
505800
2896
08:40
so you meet Briar for the first time.
171
508720
2176
08:42
And what happens?
172
510920
1336
08:44
So what happens is there is a new part,
173
512280
2296
08:46
a brand new part of the space,
174
514600
2456
08:49
becoming possible for you
in this very moment,
175
517080
4256
08:53
even without any possibility
for you to foresee this
176
521360
3856
08:57
before touching that point.
177
525240
2056
08:59
And behind this there will be
a huge set of points
178
527320
2696
09:02
that could become possible
at some later stages.
179
530040
3696
09:05
So you see the space
of the possible is very peculiar,
180
533760
2816
09:08
because it's not predefined.
181
536600
2216
09:10
It's not something we can predefine.
182
538840
2296
09:13
It's something that gets
continuously shaped and reshaped
183
541160
3376
09:16
by our actions and our choices.
184
544560
2600
09:20
So we were so fascinated
by these connections we made --
185
548120
3456
09:23
scientists are like this.
186
551600
1896
09:25
And based on this,
187
553520
2296
09:27
we conceived our mathematical formulation
for the adjacent possible,
188
555840
3216
09:31
20 years after the original
Kauffman proposals.
189
559080
3456
09:34
In our theory -- this is a key point --
190
562560
2136
09:36
I mean, it's crucially based
on a complex interplay
191
564720
3536
09:40
between the way in which
this space of possibilities expands
192
568280
4776
09:45
and gets restructured,
193
573080
1536
09:46
and the way in which we explore it.
194
574640
2496
09:49
After the epiphany of 2012,
195
577160
3856
09:53
we got back to work, real work,
196
581040
1656
09:54
because we had to work out this theory,
197
582720
1896
09:56
and we came up with
a certain number of predictions
198
584640
2416
09:59
to be tested in real life.
199
587080
1256
10:00
Of course, we need a testable framework
200
588360
2896
10:03
to study innovation.
201
591280
1456
10:04
So let me drive you
across a few predictions we made.
202
592760
4056
10:08
The first one concerns
the pace of innovation,
203
596840
2896
10:11
so the rate at which you observe novelties
in very different systems.
204
599760
4896
10:16
So our theory predicts
that the rate of innovation
205
604680
2496
10:19
should follow a universal curve,
206
607200
1936
10:21
like this one.
207
609160
1320
10:23
This is the rate of innovation versus time
in very different conditions.
208
611240
3640
10:27
And somehow, we predict
that the rate of innovation
209
615720
2616
10:30
should decrease steadily over time.
210
618360
2696
10:33
So somehow, innovation
is predicted to become more difficult
211
621080
3096
10:36
as your progress over time.
212
624200
1920
10:38
It's neat. It's interesting.
It's beautiful. We were happy.
213
626960
3536
10:42
But the question is, is that true?
214
630520
2176
10:44
Of course we should check with reality.
215
632720
1880
10:47
So we went back to reality
216
635600
2376
10:50
and we collected a lot of data,
terabytes of data,
217
638000
3136
10:53
tracking innovation in Wikipedia, Twitter,
218
641160
3336
10:56
the way in which we write free software,
219
644520
2216
10:58
even the way we listen to music.
220
646760
1640
11:01
I cannot tell you, we were
so amazed and pleased and thrilled
221
649160
3736
11:04
to discover that the same predictions
we made in the theory
222
652920
3496
11:08
were actually satisfied in real systems,
223
656440
2576
11:11
many different real systems.
224
659040
1536
11:12
We were so excited.
225
660600
1496
11:14
Of course, apparently,
we were on the right track,
226
662120
2816
11:16
but of course, we couldn't stop,
227
664960
2496
11:19
so we didn't stop.
228
667480
1496
11:21
So we kept going on,
229
669000
2096
11:23
and at some point
we made another discovery
230
671120
2056
11:25
that we dubbed "correlated novelties."
231
673200
3536
11:28
It's very simple.
232
676760
1256
11:30
So I guess we all experience this.
233
678040
1896
11:31
So you listen to "Suzanne"
by Leonard Cohen,
234
679960
3560
11:36
and this experience
triggers your passion for Cohen
235
684440
3656
11:40
so that you start frantically
listening to his whole production.
236
688120
3816
11:43
And then you realize
that Fabrizio De André here
237
691960
2296
11:46
recorded an Italian version of "Suzanne,"
238
694280
1976
11:48
and so on and so forth.
239
696280
2016
11:50
So somehow for some reason,
240
698320
1976
11:52
the very notion of adjacent possible
is already encoding the common belief
241
700320
3896
11:56
that one thing leads to another
242
704240
2560
11:59
in many different systems.
243
707720
1736
12:01
But the reason why we were thrilled
244
709480
2296
12:03
is because actually
we could give, for the first time,
245
711800
2524
12:06
a scientific substance to this intuition
246
714348
2068
12:08
and start making predictions
247
716440
1656
12:10
about the way in which
we experience the new.
248
718120
2416
12:12
So novelties are correlated.
249
720560
2320
12:16
They are not occurring randomly.
250
724320
2056
12:18
And this is good news,
251
726400
1456
12:19
because it implies
that impossible missions
252
727880
4736
12:24
might not be so impossible after all,
253
732640
2376
12:27
if we are guided by our intuition,
254
735040
3096
12:30
somehow leading us
to trigger a positive chain reaction.
255
738160
3760
12:34
But there is a third consequence
of the existence of the adjacent possible
256
742840
3496
12:38
that we named "waves of novelties."
257
746360
3536
12:41
So just to make this simple, so in music,
258
749920
2696
12:44
without waves of novelties,
259
752640
1376
12:46
we would still be listening
all the time to Mozart or Beethoven,
260
754040
6056
12:52
which is great,
261
760120
1496
12:53
but we don't do this all the time.
262
761640
1656
12:55
We also listen to the Pet Shop Boys
or Justin Bieber -- well, some of us do.
263
763320
5016
13:00
(Laughter)
264
768360
2176
13:02
So we could see very clearly
all of these patterns
265
770560
3896
13:06
in the huge amounts of data
we collected and analyzed.
266
774480
3736
13:10
For instance, we discovered
that popular hits in music
267
778240
3656
13:13
are continuously born, you know that,
268
781920
1896
13:15
and then they disappear,
still leaving room for evergreens.
269
783840
3440
13:20
So somehow waves of novelties ebb and flow
270
788120
3096
13:23
while the tides always hold the classics.
271
791240
2576
13:25
There is this coexistence
between evergreens and new hits.
272
793840
3960
13:31
Not only our theory
predicts these waves of novelties.
273
799920
2696
13:34
This would be trivial.
274
802640
1456
13:36
But it also explains why they are there,
275
804120
2896
13:39
and they are there for a specific reason,
276
807040
1976
13:41
because we as humans
display different strategies
277
809040
3216
13:44
in the space of the possible.
278
812280
1856
13:46
So some of us tend to retrace
already known paths.
279
814160
5136
13:51
So we say they exploit.
280
819320
2320
13:54
Some of us always launch
into new adventures.
281
822360
2856
13:57
We say they explore.
282
825240
1696
13:58
And what we discovered is
all the systems we investigated
283
826960
3296
14:02
are right at the edge
between these two strategies,
284
830280
3176
14:05
something like 80 percent exploiting,
20 percent exploring,
285
833480
3536
14:09
something like
blade runners of innovation.
286
837040
2680
14:12
So it seems that the wise balance,
you could also say a conservative balance,
287
840720
5216
14:17
between past and future,
between exploitation and exploration,
288
845960
4976
14:22
is already in place
and perhaps needed in our system.
289
850960
3416
14:26
But again the good news is
now we have scientific tools
290
854400
3616
14:30
to investigate this equilibrium,
291
858040
1736
14:31
perhaps pushing it further
in the near future.
292
859800
3280
14:37
So as you can imagine,
293
865360
2256
14:39
I was really fascinated by all this.
294
867640
4160
14:44
Our mathematical scheme
is already providing cues and hints
295
872920
3136
14:48
to investigate the space of possibilities
296
876080
2056
14:50
and the way in which
all of us create it and explore it.
297
878160
4016
14:54
But there is more.
298
882200
1336
14:55
This, I guess, is a starting point
of something that has the potential
299
883560
3376
14:58
to become a wonderful journey
for a scientific investigation of the new,
300
886960
4616
15:03
but also I would say
a personal investigation of the new.
301
891600
3280
15:09
And I guess this can have
a lot of consequences
302
897320
2896
15:12
and a huge impact in key activities
303
900240
2136
15:14
like learning, education,
research, business.
304
902400
5320
15:20
So for instance, if you think
about artificial intelligence,
305
908680
2896
15:23
I am sure -- I mean,
artificial intelligence,
306
911600
2136
15:25
we need to rely in the near future
307
913760
1816
15:27
more and more on the structure
of the adjacent possible,
308
915600
3816
15:31
to restructure it, to change it,
309
919440
1936
15:33
but also to cope
with the unknowns of the future.
310
921400
2320
15:36
In parallel, we have a lot of tools,
311
924400
1856
15:38
new tools now, to investigate
how creativity works
312
926280
3496
15:41
and what triggers innovation.
313
929800
1600
15:44
And the aim of all this
is to raise a generation of people
314
932080
3176
15:47
able to come up with new ideas
to face the challenges in front of us.
315
935280
3616
15:50
We all know.
316
938920
1216
15:52
I think it's a long way to go,
317
940160
2096
15:54
but the questions, and the tools,
318
942280
3056
15:57
are now there, adjacent and possible.
319
945360
3560
16:01
Thank you.
320
949720
1216
16:02
(Applause)
321
950960
4880

▲Back to top

ABOUT THE SPEAKER
Vittorio Loreto - Physicist
Vittorio Loreto is passionate about the complexity of the world around us in all its forms and he actively tries to decode it.

Why you should listen

Vittorio Loreto is a physicist at Sapienza University of Rome and faculty of the Complexity Science Hub Vienna. He is presently director of the SONY Computer Science Laboratories in Paris where he heads the team on creativity, innovation and artificial intelligence. He recently coordinated the research program dubbed KREYON, aimed at unfolding the dynamics of creativity, novelties and innovation. While theoretical modeling and data analysis are his native research tools, in the last few years he has been developing interactive tools, games, installations, to directly involve the public on the very research agenda. He created the KREYON DAYS, a new form of scientific event that tightly entangles research, learning, awareness and fun.

More profile about the speaker
Vittorio Loreto | Speaker | TED.com