Archive for the ‘Life as a scientist’ Category

Hidden gems

Tuesday, January 29th, 2013

Hidden on CQT’s website are some veritable gems which can be found among the colloquia given at CQT. All recent colloquia have been video taped, so that everyone can watch the talks online, and with the impressive list of speakers, you are bound to find plenty of talks of interest.

The colloquium is a monthly installment, where speakers have about an hour to give a more high level view of their field than during a normal talk. This makes for some really entertaining talks, and I can’t wait to watch the presentation that Eric Cornell gave last Wednesday again. In it, he managed to motivate his interest in making more precise measurements of the electron’s electric dipole moment by his desire to make supersymmetry the Paris Hilton of physics. That has gotten you interested? Unfortunately you have to wait a little bit more, because the video usually comes up about three weeks after the colloquium.

For now you can browse the colloquium archive. Last December for example, Avi Widgerson gave a highly enjoyable talk about randomness. His introduction to what randomness means from a computational scientist perspective is very interesting and the talk is easy to follow. Since it was given at CQT’s fifth birthday, the introduction of the speaker is a bit longer then usual and the talk starts for real only at the 4:30 mark of the video.

Update: Eric Cornell’s talk is now available online.

Sean Barrett

Sunday, October 21st, 2012

I was greatly saddened to hear of the death of Sean Barrett on Friday morning. It seems that a taxi he was traveling in was hit by a stolen SUV, while he was on his way to a conference in Perth. I was really shocked to hear the news.

Sean is pretty known within the quantum computing community for his work on entanglement generation and on measurement based computation. Indeed, one of his papers (on double-heralding entanglement generation) was a big influence on much of my PhD research.

Sean was a great guy, friendly, lively, and very smart. His death is a great loss to the community and my thoughts are with his family and loved ones at what must be a very difficult and painful time.

Revealing the impossible

Monday, August 27th, 2012

“Is the task impossible or does it only seem that way? The beauty is: Nobody knows.”

Someone talking about solving the Traveling Salesman Problem in polynomial time? No, though I wouldn’t blame you for thinking that. This is from a new book of wire walker extraordinaire Philippe Petit, “Cheating the Impossible: Ideas and recipes from a rebellious high-wire artist“. It is a short TED book, available as an Amazon single. Here is the related TED talk, though I found the book better than the talk (and does not take much longer to read).

The research rollercoaster

Wednesday, August 8th, 2012

In the last post, I mentioned an open problem. Last week, I solved it.

I had the idea of something to try for a few days and finally got around to it on Monday. I didn’t think it would work, but thought it would be interesting to see why it failed. By late afternoon, nothing had failed and I became very excited.

For me, this is one of the best moments in research. It is a walk along a cliff edge—the exhilirating height of discovery and danger that at any moment a hole can appear that sends you and your proof tumbling down. Sometimes I just sit on the cliff and enjoy the view—purposefully delaying writing up the proof because I know it can all come crashing down, and I want to prolong those moments of joy a little while longer.

In this case, the initial verifications passed. I worked out a small example that also checked out. I was becoming fairly convinced.

I thought about canceling my plans for the evening to write up the proof straight away. But again I thought why not wait and enjoy these moments of new proof a little while longer.

I got back home around 11pm and started in on the proof, typing everything up. As darkness lifted in the early morning hours, the proof was done and there were no bugs. For this problem, I could write a program to check the algorithm, which also worked. At this point all doubts had evaporated.

It was late. I was tired. I was just putting the final touches on the writeup, adding some references so I could send it to some colleagues.  I was googling to get the bibliographical data. Then a paper appeared in the list of search results that I had not seen before.  I clicked on it. It solved the same problem! I went to bed depressed.

That is the research rollercoaster.

When physicists on the little red dot gather

Friday, February 24th, 2012

So here it is, the meeting of all physicists on the little red dot. Or as it is officially known, the IPS Meeting 2012, this year’s edition of the annual meeting of the Institute of Physics Singapore. As you can see from the program, the IPS Meeting is closely modeled after the annual meetings of bigger physical societies such as the APS in the United States of the DPG in Germany. There are exhibitors, plenary talks, invited talks, contributed talks and plenty of space for poster sessions.

So much for the theory. But will it fly? Isn’t the local physics community to small? Aren’t the topics to widespread? Isn’t there culture too different from the States or Europe? The answer to that is a clear no. Wandering through the halls of the conference venue during the coffee and lunch breaks, you will see that there is a vivid exchange of ideas. And seeing how the conference brings together physicists from all over Singapore, you are easily let to the conclusion that the IPS Meeting has no reason to fear comparison to its bigger cousins. Except for the fact that the whole conference comfortably fits into three seminar rooms and the connecting hallways.

Irrationality of reviewing

Thursday, February 23rd, 2012

I have to finish reviewing two papers today.  Posting about the irrationality of reviewing seems like a great way to procrastinate.

Why do we review papers?  As pointed out in the boycott against Elsevier, with the big commercial publishers, reviewers are doing free work for very profitable companies.  Could it actually be that we are more likely to do reviews because they are not paid?

In Predictably Irrational, Dan Ariely distinguishes actions we undertake as part of social norms versus market norms.  When we help a friend move a sofa, we would probably be offended if he offered to pay us something at the end—this action takes place in the social realm, not the market realm.  Ariely did experiments (with college students of course) where he found subjects worked harder when asked to do a simple task as a favor to the experimenter than when they were paid some nominal amount.  Similarly, he relates the story of a daycare center that found that the tardiness of parents picking up their kids increased when the center started imposing fines for lateness.  Before the fines, parents avoided being late as a courtesy to the teachers who they were keeping from going home.  With fines this moved to a market interaction, where you could pay to be late.

Reviews largely work in the realm of social norms.  Many review requests I receive are from colleagues I know well, making them hard to turn down.  It is much easier to say no to form letters from people I do not know, and it would be similarly easier to say no if it became a (low) paid job rather than a favor.

While on the topic of reviewing irrationality, I remember an old post of Lance Fortnow that when evaluating results in a paper, sometimes less is more.  He related the story of a student who had a paper rejected from STOC, then removed one of the two main theorems and won best student paper at FOCS.

Being on a pop economics kick, I just finished reading Thinking, Fast and Slow by Daniel Kahneman, who discusses this very “less is more” phenomenon.  This book was very enjoyable, full of interesting examples of how real human behavior differs from that of the Homo Economicus usually assumed by economists.

Kahneman describes an experiment where subjects are asked to put a price on a set of dinnerware.   Set A has 24 plates and bowls.  Set B is set A with the addition of 8 cups and 8 saucers, a few of which are broken.

In single evaluation, where subjects are presented only one of set A or set B, set A was valued at $33 while set B was valued at$23.  In joint evaluation, where both sets are presented simultaneously, subjects acknowledge that set B can only be better than set A, and valued set B at $32 and set A at$30.

I think the implication for paper submission is clear: if in doubt, submit multiple versions and force your reviewers to behave rationally with joint evaluation.

An open revolution

Wednesday, February 8th, 2012

We are in the midst of an open revolution in science. This is a revolution towards doing research in the open, using online tools to share unfinished ideas and allow massive collaboration, but also an open revolt against the traditional model of publishing those finished ideas, in particular against the publishing giant Elsevier.

In his book Reinventing Discovery, Michael Nielsen (co-author of The Book in quantum information) is an enthusiastic chronicler of the rise of open science, and inspiring advocate for its success.  Some of his examples were familiar to me, like the arXiv or the polymath project.  But there were also some neat examples I had not heard of, like the Galaxy Zoo, where anyone can help classify galaxies, FoldIt, where users compete video game style to come up with low-energy configurations of proteins, or Kasparov vs. The World, where 50,000 people democratically played chess against Kasparov, doing much better than expected.

One example I was surprised not to find in the book was the stack exchange family of question and answer websites, mentioned previously on this blog.  Sites like math overflow or cs theory stack exchange, where people can ask research-level questions and expect to get informative answers in a matter of hours, have quickly become very useful research tools.

The end of Nielsen’s book is a call to arms, with suggestions of what scientists, programmers, citizens can do to develop open science, and also some fantasies of what open science might look like in the future.  One such idea I found quite cool was a sort of stack exchange meets eHarmony service.  This tool would automatically pair questions with respondents, based on the respondents expertise and questions they have successfully answered previously.  As Nielsen describes it, every morning a researcher would receive an email with a list of ten requests related to their expertise, which they could then choose to answer or not.

What I personally find the most exciting in the open science revolution is the development of open education.  The Khan academy deserves a lot of credit for innovation here, showing that students often prefer a short video that can be paused and replayed to a live lecture.  They now have over $$10^8$$ downloads and 2600 videos.

Sebastian Thrun, who co-taught the Stanford online artificial intelligence class last semester that attracted 160,000 students, recently gave a talk about the experience which I think is well worth watching.  Thrun says that this experienced changed his life, and the talk is surprisingly heartfelt and emotional.  He also has some harsh words for the current university teaching model: “Miraculously, professors teach today exactly the same way they did 1000 years ago.  University has been, surprisingly, the least innovative of all places in society.”  Nielsen also addresses this point in discussing the low rate of participation of researchers in Wikipedia.

Thrun further says that after teaching 160,000 students (of which 23,000 completed the course), he can’t go back to the old way of teaching the 200 in his normal Stanford classes.  He gave up tenure at Stanford in April of last year and now has began a startup, udacity, focused on producing online classes.  Two classes are on tap for the first offering, CS101 going from zero programming experience to programming a web search engine, and CS373, programming some of the algorithms under the hood of a driverless car.