5th HLF – Interviews with mathematics and computer science laureates: Daniel Spielman
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 49 | |
| Autor | ||
| Lizenz | Keine Open-Access-Lizenz: Es gilt deutsches Urheberrecht. Der Film darf zum eigenen Gebrauch kostenfrei genutzt, aber nicht im Internet bereitgestellt oder an Außenstehende weitergegeben werden. | |
| Identifikatoren | 10.5446/40111 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
00:00
Elektronisches ForumComputerMathematikFormation <Mathematik>Computeranimation
00:12
Elektronisches ForumNevanlinna-TheorieJSONXMLUML
00:22
Nevanlinna-TheorieNumerische MathematikTwitter <Softwareplattform>t-TestFakultät <Mathematik>MultiplikationsoperatorNeuroinformatikDifferenteFluktuation <Statistik>Gesetz <Physik>MathematikerEreignishorizontInformatikAlgorithmische ProgrammierspracheBesprechung/Interview
02:08
Nevanlinna-TheorieGüte der AnpassungFitnessfunktionKategorie <Mathematik>MereologieQuick-SortMultiplikationsoperatort-TestBesprechung/Interview
03:41
Nevanlinna-TheorieMultiplikationsoperatorRechter WinkelBesprechung/Interview
04:45
Nevanlinna-Theoriet-TestNeuroinformatikTheoremLang, SergeEinsMultiplikationsoperatorDifferenteEndliche ModelltheorieGradientProgrammierumgebungStatistische HypotheseFamilie <Mathematik>TermRückkopplungPunktUmsetzung <Informatik>MathematikerVollständiger VerbandMathematikNumerische MathematikPeer-to-Peer-NetzMailing-ListeQuick-SortTypentheorieGüte der AnpassungFormation <Mathematik>DämpfungBitInformatikArithmetisches MittelFigurierte ZahlComputerspielLeistung <Physik>Besprechung/Interview
11:16
Elektronisches ForumMathematikComputerComputeranimation
Transkript: Englisch(automatisch erzeugt)
00:24
So you were eligible to come to this, when did you get it? Oh my gosh, 2010, I think, yes, I remember now, 2010. So you've been eligible to come to all of this? Yes, I came to the second one.
00:42
It's difficult for me to come to them because if it happens, this time of year I'm usually teaching, and I usually don't travel while I'm teaching. So I came this time because I'm on leave this semester. And I came three years ago because I was on leave that semester. Have you noticed any difference between the two years?
01:01
Yes, there are many fewer mathematicians here this year, and at least among the laureates, and I don't know why that is. That's not disturbing. I'm hoping it's a random fluctuation due to the law of small numbers rather than a trend. We will see. I mean, I've noticed that it's many of the same people who are here.
01:20
So it might just be that this event attracts certain people who are outgoing or who like coming to Germany or Heidelberg. I'm not sure. And this year is the first year they've had the ACM Prize, which used to be the infrastructure. That's right. So that brings a few more computer scientists, or two more computer scientists this year, I think. But I think it's only two of them.
01:42
But you've come back, so obviously you like it. Tell me some of the things you get from that. Well, OK, so the first reason I came back is I... Wow, OK. So the first reason... The first reason I came back is I talked to many people who came here as students or postdocs or young faculty
02:03
to ask them what they thought, whether or not they liked it. And a lot of the people I talked to said that they thought they got a lot out of it. So what I wanted to do was confirm that coming here was actually providing something valuable to all of the students, rather than just coming to spend time in a beautiful city and ride on boats and eat good food.
02:23
But it seemed that, yes, for a lot of the young scientists here, it's actually a lot of them find it very useful and found it very inspirational. So that's why I come back. And I assume that a lot of them come up to you and they want to sort of get knowledge about what to do next.
02:41
What do you tell them? What sort of questions do you get and what do you tell them? I don't think anyone has asked me what to do next. A lot of people tell me about what they're doing and I try to give them some advice or my reactions or what it sounds like to me. Some of them ask me what I'm working on
03:01
and I tell them about some of the crazier things I'm trying. Most of the things I'm working on fit in that category of being very, things that are very unlikely to succeed. I try to, I do try to impart sort of my generic advice, which is that, one, I try to only work on things
03:22
that I'm really passionate about. There are many times during my career when I've tried to work on things that I thought I should do, but if I wasn't passionate about, it just never succeeded. Part of that was just because I think I just didn't spend enough time thinking about those things and would go back to the things I was passionate about, even if they weren't necessarily sensible.
03:43
I also try to warn them about how much I fail and reassure them that the amount that they fail is yes, it's the same. Right, I mean, as I explained to them, I work on things for months and months and months and probably only have two good ideas a year and those are really good,
04:03
but most of what I do doesn't work and so right, it's a career that you have to really have, I don't know if it's a stomach for failure or really enjoy what you're doing along the way or I also describe it's like suffering, almost like being a gambler. So there are many times other than those two times a year when I think I have a great idea
04:21
and I'm incredibly excited and so happy and I can't even sleep sometimes, I'm so excited. And my wife will tell me, oh, you know, just go to sleep, you'll find the mistake in the morning. She's usually right. Most of these ideas don't work, but I think the thrill of having an idea that almost works until I find the mistake
04:40
somehow keeps me going. That was gonna be my next question actually, is how do you deal with disappointment and it sounds like persistence plays a big part in it. Yeah, getting back to work helps. It helps to be working on problems that I enjoy or I have to enjoy working on the problem and occasionally, rarely, but occasionally just giving up
05:01
and working on a different problem. I mean, okay, I should say, I usually have more than one problem in mind on any given day or any given week, partially so that when I run out of ideas on one, I can start working on another. And this is the way I've functioned since grad school. I find if I'm just working on one thing, then I really will become sort of depressed and down
05:23
when that starts failing and I don't know what to do next. It's good if I at least have one other problem to start thinking about. Have you always been this way? This is what I've done sort of my whole career. I mean, at least since I started graduate school, I've always been trying to work on more than one thing. And graduate school is a good time to do that
05:42
because in graduate school, you have a lot of time. So I'd say, you know, graduate school, that I could work on one problem the first six hours of the day and then have lunch and work on another problem the next six hours of the day. But you're putting in 12 hours. Yes, I used to put in quite a bit of time in graduate school.
06:00
And it's joking a little bit because I also had to study for courses and things like that. But graduate school is a good time to put in a lot of time. You don't get that opportunity again. What other things do you think, because a lot of the people here are 22 to 27 or so, and I've heard, especially for mathematicians, that's such a crucial time. What advice would you give people at that age,
06:22
what other advice would you give for looking ahead five years, you know? Oh, gosh. I mean, I don't know. I mean, I say, right, it's a very good time to work hard, but you also can't ignore the rest of your life. If you're lonely, you should start putting some of that brain power to use
06:41
on meeting someone you'll fall in love with. It might sound like a cynical way to think of it, but I think that's a good way to break through to someone who's spending a lot of time on math or computer science. So who are your mentors at that time? So I've been fortunate to have had a number of excellent mentors.
07:00
When I was an undergraduate, I largely apprenticed myself, at least in mathematics, to Serge Lang. He was very committed to trying to train undergraduates to go on to become mathematicians, and he, conversely, was willing to put in immense hours training us, so I and many of my peers decided, you know, we would take every course he taught
07:20
and learn from him as much as possible. I was also very fortunate when I was an undergraduate to be able to do research in computer science. I worked with a professor at Yale named Richard Beagle, who gave me a research problem, and then later we wrote a few papers together and managed to keep working. Again, when I was in graduate school,
07:41
I had some amazing role models. Giancarlo Rota was one of the great combinatorialists, and he was, I was at MIT for graduate school, and he was one of the people who I really tried to emulate, partially because he'd had a huge impact, both intellectually, but also personally.
08:00
It felt like everyone around him was happy, and that was something that I wanted to somehow emulate and create as an environment for me. Also, my thesis supervisor, Mike Sipser, was terrific in terms of helping me think about what to work on, listening to all of my crazy ideas, and he would listen to an hour for something,
08:24
and sometimes he'd give me useful feedback. Sometimes he wasn't sure what he said that it was useful, but it was often he would say a lot that was useful, and giving me just thoughts about strategizing how to work my career and how to be a professor. But one of my main goals when I was in graduate school was to learn how to emulate the professors I saw.
08:44
So a lot of them were my mentor, whether they knew it or not, meaning I would ask them questions, not necessarily because I wanted to know the answer, but because I wanted to learn how they reacted when they were asked questions, so that when I am faced with a problem, I will often think to myself,
09:02
like, what would Danny Kleitman do in this situation? Or what would Tom Leighton do in this situation? And the reason I can think that is because I asked them problems and saw what they did in different situations, and then it's just one of the ways I try to come up with inspiration is I try to pretend I'm someone else and get into character, then try to think about what they do.
09:21
What do you think your style is in mentoring people? I try as much as possible to figure out what is the right style for each student, and it's often different. So I mean, I've tried many things. There's some students who I just give many, many papers to read, and I ask them to come and explain them to me and tell me what their reactions are,
09:42
and this is one, a good point for starting conversations. And helps them learn about things that are of interest to me, and maybe helps them see when they're really excited about a paper, we'll ask questions and try to figure out what to do next. There are other students who I work with very differently. There are some students who I've worked with
10:01
where I do a lot of computational experiments and use them to make conjectures, and then I try to get them to work on those conjectures. I mean, one of my students said it felt rather backwards that I was doing experiments and he was proving theorems. But that worked. There are other students I've had where they've really been the ones
10:20
to come up with the conjectures, and I've been the one to put into time to try and figure out how to prove them. So every relationship really is different. Anything else you've heard about the HLF or mentoring or anything all the way through? I think HLF is actually a very interesting place
10:41
for students to come and be exposed to many different successful role models, because I think it is good for them to see that as a bunch we are fairly different, and that should help them get in mind that they don't have to be any one particular way to succeed. There are all sorts of people who do
11:01
and all sorts of different personality types and different people have different talents, and I think the people who succeed are those who figure out how to get maximal use out of the few talents they have. Thank you for your good final statement. Thank you so much. You're welcome.