aporetic - David Molnar
[Most Recent Entries]
[Calendar View]
[Friends]
Below are the 20 most recent journal entries recorded in
David Molnar's LiveJournal:
[ << Previous 20 ]
| Tuesday, June 30th, 2009 | | 4:17 pm |
computer science research on cap and trade?
There's a lot of attention in the economics of computer science community on auctions or other economic mechanism design, often with ads in mind. What is under way with respect to mechanism design for cap and trade of carbon emissions? These markets are being set up now, so it seems like this is a great time for research. For that matter, how do existing proposals for carbon cap and trade compare to the kind of mechanisms studied in, say, ACM Electronic Commerce? This post brought to you by the recently passed (barely!) federal greenhouse gas cap and trade bill. | | Sunday, June 14th, 2009 | | 8:56 pm |
| | Wednesday, June 10th, 2009 | | 1:34 pm |
Seminar on reputations, self-control, and the prefrontal cortex at Duke
Via Vincent Conitzer's announcement list. I won't be able to make it, but looks like the seminar will have interesting things to say about where reputation comes from (and by extension how to model human behavior?). --- Daniel Schunk (The Institute for Empirical Research in Economics at the University of Zurich) to speak on "Reputation formation, self-control, and the prefrontal cortex." June 23, 2009 ~ 11:00 a.m. ~ Erwin Mill room A103 Abstract: Reputation formation pervades human social life, and the ubiquity of reputation mechanisms distinguishes humans from all other living species. Previous studies demonstrated that reputations for being trustworthy and generous play a key role in cooperation among genetically unrelated individuals. We know little, however, about the neural underpinnings of this important social mechanism. Here we show that disruption of the right, but not the left, dorsolateral prefrontal cortex (DLPFC) with low-frequency repetitive transcranial magnetic stimulation (rTMS) diminishes subjects' ability to build a favorable reputation. This effect occurs even though subjects' ability to behave altruistically in the absence of reputation incentives remains intact, and even though they are still able to recognize both the fairness standards necessary for acquiring and the future benefits of a good reputation. Thus, subjects with a disrupted right DLPFC no longer seem to be able to resist the temptation to defect, even though they know that this has detrimental effects on their future Here we show that disruption of the right, but not the left, dorsolateral prefrontal cortex (DLPFC) with low-frequency repetitive transcranial magnetic stimulation (rTMS) diminishes subjects' ability to build a favorable reputation. This effect occurs even though subjects' ability to behave altruistically in the absence of reputation incentives remains intact, and even though they are still able to recognize both the fairness standards necessary for acquiring and the future benefits of a good reputation. Thus, subjects with a disrupted right DLPFC no longer seem to be able to resist the temptation to defect, even though they know that this has detrimental effects on their future reputation. This suggests an important dissociation between the knowledge about one's own best interests and the ability to act accordingly in social contexts. These results link findings on the neural underpinnings of self-control and temptation with the study of human social behavior, and they may help explain why reputation formation remains less prominent in most other species with less developed prefrontal cortices. ----- This talk is co-sponsored by Duke's Center for Neuroeconomic Studies, Social Science Research Institute, and the Duke Institute for Brain Sciences. All are welcome. Please circulate this announcement widely. | | Wednesday, June 3rd, 2009 | | 12:15 pm |
KLEE open sourced
The KLEE tool by Cristian Cadar, Daniel Dunbar, and Dawson Engler is now open source. This tool achieved impressive coverage results on a wide array of unix utilities, and now you can try it on your own code. Check out the really neat Getting Started page here: http://klee.llvm.org/GetStarted.htmlThis is an exciting day! | | Sunday, May 3rd, 2009 | | 5:17 pm |
decision procedure for reduction from set disjointness?
Does anyone know if there is a decision procedure of the following form: Input: A specification of a decision problem P. Decide: Does there exist a streaming reduction from the SET-DISJOINTNESS problem to P? I'm being vague of course in what I mean by "specification of a decision problem." A restricted specification language is OK, probably needed to pull off something like this. A decision procedure that says "I don't know" is OK, too. Anything of this flavor would be interesting, doesn't have to be SET-DISJOINTNESS as the problem. I'm asking because some of the work I've done involves inferring types for program values in traces of program execution. One of the observations I've been kicking around is that "efficient" algorithms for analyzing dynamic program executions should work in the streaming model, i.e. you see each instruction once, then it's gone, since for large programs you don't really want to store every instruction of the trace and do random access. SET-DISJOINTNESS is a problem for which we know no algorithm in this model exists which takes constant space. So it'd be interesting to say "here's a property we want to check or infer on traces, now can you tell me if there is evidence that no streaming algorithm exists?" | | Thursday, April 23rd, 2009 | | 9:19 am |
| | Saturday, April 18th, 2009 | | 7:35 pm |
| | Monday, March 16th, 2009 | | 11:23 pm |
Jaw dropping moment of the day!
Title: Fully Homomorphic Encryption Using Ideal Lattices Speaker: Craig Gentry, Stanford University Time/Place: 11 am, 18 March, Wozniak Lounge [Ed. note: 4th floor, Soda Hall, UC Berkeley] Abstract: We propose a fully homomorphic encryption scheme -- i.e., a scheme that allows one to evaluate circuits over encrypted data without access to the decryption function. First, we provide a general preliminary result -- that, to construct an encryption scheme that permits evaluation of arbitrary circuits, it suffices to construct an encryption scheme that can evaluate (slightly augmented versions of) its own decryption circuit; we call such a scheme bootstrappable. Next, we provide a bootstrappable public key encryption scheme using ideal lattices. | | Saturday, March 14th, 2009 | | 5:04 pm |
Art reception tonight in San Francisco on "Algorithms"
Algorithmia: An exhibition about problem solving.http://rootdivision.org/031409.html"Just as the algorithms of nature constantly work in flux to solve and create problems, so does the human use of algorithmic processes towards advancement. The science fictions of yesterday are rapidly becoming the realities that shape our everyday. From iPods shuffling our favorite songs and videos, to web applications that allow us to do everything remotely online, to the “smart bombs” that target enemies of the corporate world, we are surrounded by the good, the bad, and the ugly sides of computational progress. Algorithmia consists of both physical and web-based work, presenting diverse procedures for problem solving, and explores the ways these methods, in turn, program us." Reception: Saturday, March 14th, 7 - 10 pm Sliding Scale Suggested Donation: $2-$20 Exhibition Dates: March 11-28, 2009 Gallery Hours:Wednesdays- Saturdays, 12-4 pm (or by appointment) ROOT DIVISION GALLERY 3175 17th Street (at S. Van Ness) San Francisco, CA 94110 Curated By: Bryan Hewitt, Vita Mei Hewitt, and Lauren Scime Featured Artists: Alan Bigelow David Sanchez Burr Elizabeth Deters* Ethan Ham Bryan Hewitt Vita Mei Hewitt Ryan Jones* Emmanuelle Namont Kouznetsov Kunsole* Conrad M. Meyers II* Lauren Scime* Luther Thie & Carlos Castellanos Myriam Thyes Fernando Velázquez, Bruno Favaretto & Francisco Lapetina *Root Division Resident Artist ABOUT ROOT DIVISION: Root Division is an arts and arts education non-profit located in the Mission District of San Francisco. Root Division's mission is to improve appreciation and access to the visual arts by connecting personal inspiration and community participation. We provide subsidized studio space to working artists in exchange for their service in creating shared learning opportunities for the community. Artists develop creatively and professionally by teaching art to underserved youth, leading adult education classes, and producing exhibitions that showcase local emerging artwork. By combining multiple opportunities for creative exchange, Root Division cultivates an artistic ecosystem that enriches life throughout the Bay Area. | | Friday, February 20th, 2009 | | 2:53 am |
Micro-posts
A kind "systems grad student" at Michael Mitzenmacher's blog pointed out that I have not posted hardly any public entries recently. This is true. I have had a few ideas for public posts over the last few months, but I've never quite gotten around to writing them up. The more time I take on each one, the more I want to spend polishing, which of course never converges. So I thought I'd try something different - post these sketches of posts, and let people say which of them, if any, sound like they are worth expanding. ( notes ) Current Music: Iambia - Helluci Nation | | 2:44 am |
Cyborg beetles at Berkeley on Wednesday
Every now and then in the elevator at Cory Hall (the Berkeley electrical engineering building), I see an ad for research assistants to work on the "Cyborg Beetle" project. The ad caught my eye because it mentions that experience with RFID is a plus. Here's a link with a picture of a prototype cyborg beetle: http://www.eecs.berkeley.edu/Research/Projects/Data/105682.htmlNow in my e-mail, I find that next Wednesday, 25 February, there will be a talk on the state of the art in cyborg beetles. Unfortunately, I'll be out of town, but maybe someone reading this can make it and let us all know how it went! "Wireless Flight Control of Cyborg Beetles: Building Man-made Interfaces to Living Systems" Michel Maharbiz [Assistant Professor of Electrical Engineering and Computer Sciences, UC Berkeley] 12:00 p.m., Wednesday, February 25 290 Hearst Memorial Mining Building, Dado and Maria Banatao Conference Room, UC Berkeley Campus http://www.citris-uc.org/events/RE-Feb25 Current Music: Rotersand - Rushing | | Saturday, October 4th, 2008 | | 1:59 am |
What does knot theory have to do with P^#P != NP ?
I didn't know, but Michael H. Freedman has an answer - by assuming that the complexity class P^#P is not equal to NP, you can prove a new theorem in knot theory! Complexity Classes as Mathematical Axioms M. Freedman (Submitted on 30 Sep 2008) Abstract: Treating a conjecture, P^#P != NP, on the separation of complexity classes as an axiom, an implication is found in three manifold topology with little obvious connection to complexity theory. This is reminiscent of Harvey Friedman's work on finitistic interpretations of large cardinal axioms. http://arxiv.org/abs/0810.0033 Current Music: Crystal Castles - Air War | | Wednesday, September 24th, 2008 | | 2:27 pm |
Talk today on random SAT formulas
[Thanks to David Aldous for forwarding this to a mailing list I'm on.] Probability Seminar: Wednesday September 24, 3.10 - 4.00pm. Room 330 Evans Hall Random satisfiable formulas above the satisfiability threshold. Danny Vilenchik (UC Berkeley) k-CNF formulas with m clauses over n variables show a phase transition phenomenon in the following aspect. There exists d=d(k,n) such that almost all formulas with m/n>d are not satisfiable whereas most formulas with m/n < d are. While random k-CNFs below the threshold received much attention in recent years, above-threshold distributions over satisfiable k-CNFs were far less studied. One possible reason is that it is not clear how to define such distributions in a natural way, while keeping the problem approachable in some sense. In this talk we will survey recent developments in this area. We shall concentrate on three distributions: the planted k-SAT distribution, the uniform distribution over satisfiable k-CNF formulas (in the regime m/n>d(k,n)), and an "on-line" version of the uniform distribution. In all cases we are able to show that unlike the typically complicated structure of the solution space of below-threshold formulas, the above threshold formulas have a simple, basically single-solution structure. We also present some algorithmic ideas that are useful for solving certain clause-density regimes of these distributions. Talk based on joint works with: Amin Coja-Oghlan, Michael Krivelevich, Uriel Fiege, Abie Flaxman, and Benjamin Sudakov. | | Saturday, August 16th, 2008 | | 3:15 am |
Mini-review of "The Artist and the Mathematician"
Full title: _The Artist and the Mathematician: The Story of Nicolas Bourbaki: the Genius Mathematician Who Never Existed_ by Amir D. Aczel. I picked this up because I am curious to find exemplars of good expository writing on mathematical topics for a mass audience. After all, someday I will have to finally figure out how to explain to my parents and others "what is it that you do?" (My current software testing work I think I've figured out to explain as "breaking stuff before other people do", but hey, at some point there is crypto to do as well.) Unfortunately, while entertaining, this book falls short of what I wanted. The beginning of the book follows the personal biographies of mathematicians, such as Alexandre Grothendieck and Andre' Weil, who later became associated with Bourbaki. This part is interesting in its own right. After reading of the difficulties they encountered during the tumultuous years leading up to and including the Second World War, I will feel guilty ever complaining about my working situation again. Then, the author paints a stirring picture of the formation of the Seminaire Bourbaki and what it meant to Mathematics in France and later the world. Here I learned several tidbits of which I had not been aware, such as the citation for Bourbaki's first ever publication! The author also does an admirable job of bringing in connections between the mathematics of Bourbaki and the artistic movements in the post-war era, such as Oulipo. Coming up through an American university, I tend to be affected by C.P. Snow's "Two Cultures" viewpoint -- and even though I personally know several examples of people who excel in both "Cultures", I tend to think of them as entirely distinct. The idea of using a rule to generate poetry semi-randomly just seems like a category error to my eyes, which is why it is all the more exciting to read about it. There is also a chapter, regrettably brief, on connections between the Bourbaki programme and the linguistics of Roman Jakobson. The main issues I had with the book, however, are the wildly varying levels of mathematical maturity assumed, and with the last part of the book. As to maturity, the author cannot seem to make up his mind whether the reader is wholly unfamiliar with mathematics, or whether the reader has at least a passing acquaintaince with material that might be seen in an undergraduate curriculum. Thus we have a very vague handwaving about the properties of a group (irritating to this reader, but perhaps soothing to others), juxtaposed with a discussion of topology (p.103 in the Thunders Mouth Press edition) which uses the term "open sets" without defining what open sets actually are. Overall, I would have preferred more exposition of the mathematics itself, even at the cost of some of the intriguing personal details. What is present is hardly satisfying. Of course, if I really want the full details, I could always go read Bourbaki directly..! Finally, the last part of the book deals with Grothendieck's seclusion and the effective dissolution of Bourbaki. Here the author turns out to be a virulent Grothendieck partisan. According to him, Bourbaki failed because it did not follow the supreme generality of Grothendieck's ideas, and without Grothendieck, the group had no ability to innovate. This may be true. The author certainly has a better perspective on this than I do. Unfortunately, more space is spent claiming the great generality and power of Grothendieck's ideas rather than demonstrating and explaining such ideas. Again, I was looking for a more careful exposition of the mathematics itself, rather than a discussion of the controversies surrounding it. Overall, I am glad I read the book -- it provides me with a wealth of pointers for future reading. (Who knew, for example, that Claude Levi-Strauss's laws of kinship can be obtained using group theory? Not me.) As an example of mathematical exposition, unfortunately, it is lacking. I could recommend the book to non-mathematical friends of mine, but only after warning them that there will be the occasional passage they may not "get" and that's perfectly OK. | | Thursday, June 12th, 2008 | | 11:00 am |
| | Wednesday, June 4th, 2008 | | 11:08 am |
| | Saturday, May 31st, 2008 | | 1:21 am |
Google News of the day
I was procrastinating earlier today, and I clicked on the "Elections" tab in Google News. Notice anything?  Seems to be gone now. What's going on is that the Houston Chronicle now has a section for commenters to establish personal blogs. One of these commenters has a sense of humor...which Google News promptly found and reported. (By the way, if you don't get it, wikipedia is here to help.) | | Sunday, May 25th, 2008 | | 5:25 pm |
Moving along
Passed my last class. Now all I need to do is form a dissertation committee and write a thesis. In the meantime, via Stephen Toulouse, I have found that the Mars Phoenix has a twitter: http://twitter.com/MarsPhoenixNo idea if this is a NASA initiative or just some person following the project, but it's fun nontheless. At least, when Twitter is up. Current Music: theSTART - Wartime! (It's Time 2 Go Now) | | Monday, May 19th, 2008 | | 3:34 am |
| | Friday, May 16th, 2008 | | 10:15 pm |
|
[ << Previous 20 ]
|
![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small}
ephermata's journal