Tag Archives: design-patterns

Items of some interest:

Posted on by

These are my recent Pin​board​.in links:

  • Jonathan Lethem’s ‘Neote­nous Aes­thetic” — Mis­ter Bit — Wired​.it

    ‘Dur­ing the brief, but very inter­est­ing Q&A ses­sion, Lethem argued that inter­net cul­ture brought the “closet into the open”, that is, it gave ephemera, triv­i­al­i­ties, and every­day activ­i­ties “A new kind of vis­i­bil­ity”. “Peo­ple have always been pro­duc­ing weird stuff and have always been engag­ing in arcane activ­i­ties,” Lethem remarked. “What is really new is the fact the now we can see it. We can see it all. We can quan­tify what we do — or not do — online.” Lethem men­tioned the uncanny abil­ity to track, in real time, “how many books I am not sell­ing on Ama­zon”. “Real­ity has acquired a new level of mea­sur­a­bil­ity”. “The activ­i­ties we per­form in our dig­i­tal age are not nec­es­sar­ily new. What is new is that. We. Can. See. Them. All.”.’

    one-​​measures-​​a-​​circle inter­per­me­ation access local­ism

  • Sex, Oil, and Video­tape | Mother Jones

    “Loom­ing over Saylor’s con­fronta­tion with Bolen­baugh was the EPA’s Sep­tem­ber 27 cleanup dead­line, and it appears that Enbridge and its con­trac­tors were feel­ing the pres­sure as it drew near. In early Sep­tem­ber, after the Michi­gan Mes­sen­ger pub­lished its exposé on the use of undoc­u­mented work­ers by Hall­mark Indus­trial, another group of work­ers employed by a dif­fer­ent Enbridge con­trac­tor came for­ward with detailed sto­ries of how they had been instructed to con­ceal oil at the same site. Work­ers would land on an island, they said, remove all veg­e­ta­tion, and then lay out absorbent pom-​​poms, all per EPA reg­u­la­tions. But once the top layer of oil was absorbed, they were instructed to rake dirt over the area to make it appear as though it had been dug out. One worker described his super­vi­sor show­ing him the process step-​​by-​​step, con­clud­ing with sprin­kling a thin layer of dirt on top. “He said, ‘There, now they can’t see it. It is clean,’” the worker told the Mes­sen­ger. Another worker described being told to cover pock­ets of oil with leaves and sticks. As a last step, such areas were cor­doned off with cau­tion tape.”

    oil­spill Kala­ma­zoo local whistle­blower

  • [1204.4200] Dis­crete Dynam­i­cal Genetic Pro­gram­ming in XCS

    “A num­ber of rep­re­sen­ta­tion schemes have been pre­sented for use within Learn­ing Clas­si­fier Sys­tems, rang­ing from binary encod­ings to neural net­works. This paper presents results from an inves­ti­ga­tion into using a dis­crete dynam­i­cal sys­tem rep­re­sen­ta­tion within the XCS Learn­ing Clas­si­fier Sys­tem. In par­tic­u­lar, asyn­chro­nous ran­dom Boolean net­works are used to rep­re­sent the tra­di­tional condition-​​action pro­duc­tion sys­tem rules. It is shown pos­si­ble to use self-​​adaptive, open-​​ended evo­lu­tion to design an ensem­ble of such dis­crete dynam­i­cal sys­tems within XCS to solve a num­ber of well-​​known test problems.”

    genetic-​​programming learning-​​classifier-​​systems representation-​​theory design-​​patterns boolean-​​networks nudge-​​targets nice

  • Why Is Darwin’s Tan­gled Bank Tan­gled? : 13.7: Cos­mos And Cul­ture : NPR

    Sad to hear him still phras­ing this sim­ple truth so obscurely: Not “Because, on the scale of mol­e­c­u­lar bind­ing site recog­ni­tion, say a few tens of angstroms in length, height and width and sev­eral other fea­tures such as polar­ity, van-​​der-​​Waal forces, and so on, there are far fewer effec­tively dif­fer­ent mol­e­c­u­lar shapes than there are kinds of mol­e­cules.“ … but “Because there are fewer sto­ries than there are facts.”

    oh-​​stu pragmatism-it-ain’t philosophy-​​of-​​science

  • Math Notes | Futil­ity Closet

    So for finite sequences of dig­its, which sequences are such that the most right-​​truncated sub­strings are prime? Which are such that the most right-​​repeating exten­sions are prime?

    nudge-​​targets number-​​theory indirect-​​link

  • Home — Scal­able and Mod­u­lar Archi­tec­ture for CSS

    “I’ve been ana­lyz­ing my process (and the process of those around me) and fig­ur­ing out how best to struc­ture code for projects on a larger scale. What I’ve found is a process that works equally well for sites small and large. Learn how to struc­ture your CSS to allow for flex­i­bil­ity and main­tain­abil­ity as your project and your team grows.”

    css tuto­r­ial best-​​practices graphic-​​design via-​​trek

  • [1204.3678] Crowd Mem­ory: Learn­ing in the Collective

    “Crowd algo­rithms often assume work­ers are inex­pe­ri­enced and thus fail to adapt as work­ers in the crowd learn a task. These assump­tions fun­da­men­tally limit the types of tasks that sys­tems based on such algo­rithms can han­dle. This paper explores how the crowd learns and remem­bers over time in the con­text of human com­pu­ta­tion, and how more real­is­tic assump­tions of worker expe­ri­ence may be used when design­ing new sys­tems. We first demon­strate that the crowd can recall infor­ma­tion over time and dis­cuss pos­si­ble impli­ca­tions of crowd mem­ory in the design of crowd algo­rithms. We then explore crowd learn­ing dur­ing a con­tin­u­ous con­trol task. Recent sys­tems are able to dis­guise dynamic groups of work­ers as crowd agents to sup­port con­tin­u­ous tasks, but have not yet con­sid­ered how such agents are able to learn over time. We show, using a real-​​time gam­ing set­ting, that crowd agents can learn over time, and ‘remem­ber’ by pass­ing strate­gies from one gen­er­a­tion of work­ers to the next, despite high turnover rates in the work­ers com­pris­ing them. We con­clude with a dis­cus­sion of future research direc­tions for crowd mem­ory and learning.”

    crowd­sourc­ing learn­ing agent-​​based collective-​​intelligence mem­ory nudge-​​targets

  • [0911.1582] Manip­u­lat­ing Tour­na­ments in Cup and Round Robin Competitions

    “In sports com­pe­ti­tions, teams can manip­u­late the result by, for instance, throw­ing games. We show that we can decide how to manip­u­late round robin and cup com­pe­ti­tions, two of the most pop­u­lar types of sport­ing com­pe­ti­tions in poly­no­mial time. In addi­tion, we show that find­ing the min­i­mal num­ber of games that need to be thrown to manip­u­late the result can also be deter­mined in poly­no­mial time. Finally, we show that there are sev­eral dif­fer­ent vari­a­tions of stan­dard cup com­pe­ti­tions where manip­u­la­tion remains polynomial.”

    algo­rithms eco­nom­ics game-​​theory nudge-​​targets

  • Intro­duc­ing the Jour­nal of Dig­i­tal Human­i­ties — ProfHacker — The Chron­i­cle of Higher Education

    “If the con­tents of the inau­gural issue—which range from an essay argu­ing that human­ists need to under­stand and inter­pret quan­ti­ta­tive data to a review of the Word­Seer text analy­sis tool—fall out­side your usual schol­arly domain, then cer­tainly the journal’s edi­to­r­ial and pub­lish­ing appa­ra­tus will piqué your interest. As Dan Cohen explained in a sep­a­rate blog post, the jour­nal oper­ates under the model of catch­ing the good—of find­ing sub­stan­tive and valu­able dig­i­tal human­i­ties work “in what­ever for­mat, and wher­ever, it exists.” Blogs, pod­casts, Twit­ter con­ver­sa­tions, slideshows, and so on, these are all venues in which sig­nif­i­cant and, though I hate to use such an ungainly word, impact­ful work is being done. The reg­u­lar and guest edi­tors “catch” this work, and then pro­vide lay­ers of eval­u­a­tion and review before it appears in JDH.”

    digital-​​humanities jour­nal to-​​read two-​​cultures-​​only-​​one-​​of-​​which-​​can-​​write

  • [1005.4159] The Com­plex­ity of Manip­u­lat­ing $k$-Approval Elections

    “An impor­tant prob­lem in com­pu­ta­tional social choice the­ory is the com­plex­ity of unde­sir­able behav­ior among agents, such as con­trol, manip­u­la­tion, and bribery in elec­tion sys­tems. These kinds of vot­ing strate­gies are often tempt­ing at the indi­vid­ual level but dis­as­trous for the agents as a whole. Cre­at­ing elec­tion sys­tems where the deter­mi­na­tion of such strate­gies is dif­fi­cult is thus an impor­tant goal. …”

    vot­ing game-​​theory design-​​patterns mechanism-​​design nudge-​​targets

  • [0903.1147] Tetravex is NP-​​complete

    “Tetravex is a widely played one per­son com­puter game in which you are given $n^2$ unit tiles, each edge of which is labelled with a num­ber. The objec­tive is to place each tile within a $n$ by $n$ square such that all neigh­bour­ing edges are labelled with an iden­ti­cal num­ber. Unfor­tu­nately, play­ing Tetravex is com­pu­ta­tion­ally hard. More pre­cisely, we prove that decid­ing if there is a tiling of the Tetravex board is NP-​​complete. Decid­ing where to place the tiles is there­fore NP-​​hard. This may help to explain why Tetravex is a good puz­zle. This result com­pli­ments a num­ber of sim­i­lar results for one per­son games involv­ing tiling. For exam­ple, NP-​​completeness results have been shown for: the offline ver­sion of Tetris, KPlumber (which involves rotat­ing tiles con­tain­ing draw­ings of pipes to make a con­nected net­work), and short­est slid­ing puz­zle prob­lems. It raises a num­ber of open ques­tions. For exam­ple, is the infi­nite ver­sion Turing-​​complete? How do we gen­er­ate Tetravex prob­lems which are truly puz­zling as ran­dom NP-​​complete prob­lems are often sur­pris­ing easy to solve? Can we observe phase tran­si­tion behav­iour? What about the com­plex­ity of the prob­lem when it is guar­an­teed to have an unique solu­tion? How do we gen­er­ate puz­zles with unique solutions?”

    mathematical-​​recreations computational-​​complexity algo­rithms nudge-​​targets

  • [1204.4286] Fair Allo­ca­tion With­out Trade

    “We con­sider the age-​​old prob­lem of allo­cat­ing items among dif­fer­ent agents in a way that is effi­cient and fair. Two papers, by Dolev et al. and Ghodsi et al., have recently stud­ied this prob­lem in the con­text of com­puter sys­tems. Both papers had sim­i­lar mod­els for agent pref­er­ences, but advo­cated dif­fer­ent notions of fair­ness. We for­mal­ize both fair­ness notions in eco­nomic terms, extend­ing them to apply to a larger fam­ily of util­i­ties. Not­ing that in set­tings with such util­i­ties effi­ciency is eas­ily achieved in mul­ti­ple ways, we study notions of fair­ness as cri­te­ria for choos­ing between dif­fer­ent effi­cient allo­ca­tions. Our tech­ni­cal results are algo­rithms for find­ing fair allo­ca­tions cor­re­spond­ing to two fair­ness notions: Regard­ing the notion sug­gested by Ghodsi et al., we present a polynomial-​​time algo­rithm that com­putes an allo­ca­tion for a gen­eral class of fair­ness notions, in which their notion is included. For the other, sug­gested by Dolev et al., we show that a com­pet­i­tive mar­ket equi­lib­rium achieves the desired notion of fair­ness, thereby obtain­ing a polynomial-​​time algo­rithm that com­putes such a fair allo­ca­tion and solv­ing the main open prob­lem raised by Dolev et al.”

    eco­nom­ics game-​​theory fair­ness algo­rithms phi­los­o­phy design-​​patterns

  • Why is Esti­mat­ing so Hard? | 8th Light

    “It turns out that we don’t know the pro­ce­dure. We haven’t got any clue to just how dif­fi­cult the pro­ce­dure is. We aren’t com­put­ers. We don’t fol­low pro­ce­dures. And so com­par­ing the com­plex­ity of the man­ual task, to the com­plex­ity of the pro­ce­dure is invalid. This is one of the rea­sons that esti­mates are so hard, and why we get them wrong so often. We look at a task that seems easy and esti­mate it on that basis, only to find that writ­ing down the pro­ce­dure is actu­ally quite intri­cate. We blow the esti­mate because we esti­mate the wrong thing.”

    esti­ma­tion agile-​​practices philosophy-​​of-​​engineering man­age­ment self-​​definition plan­ning

  • [1204.4374] Higher Order City Voronoi Diagrams

    “We inves­ti­gate higher-​​order Voronoi dia­grams in the city met­ric. This met­ric is induced by quick­est paths in the L1 met­ric in the pres­ence of an accel­er­at­ing trans­porta­tion net­work of axis-​​parallel line segments. …”

    computational-​​geometry algo­rithms voronoi-​​diagrams diver­sity network-​​theory nudge-​​targets

  • Topic mod­el­ing made just sim­ple enough. | The Stone and the Shell

    “Com­puter sci­en­tists make LDA seem com­pli­cated because they care about prov­ing that their algo­rithms work. And the proof is indeed brain-​​squashingly hard. But the prac­tice of topic mod­el­ing makes good sense on its own, with­out proof, and does not require you to spend even a sec­ond think­ing about “Dirich­let dis­tri­b­u­tions.” When the math is approached in a prac­ti­cal way, I think human­ists will find it easy, intu­itive, and empow­er­ing. This post focuses on LDA as short­hand for a broader fam­ily of “prob­a­bilis­tic” tech­niques. I’m going to ask how they work, what they’re for, and what their lim­its are.”

    text-​​processing clas­si­fi­ca­tion algo­rithms lovely two-​​cultures-​​only-​​one-​​of-​​which-​​can-​​write

  • Math­e­mati­cians are Giraffe Hunters by Barry Mazur | berfrois

    “No won­der life (i.e., the thing that my once 10-​​year old niece referred to as “the thing that isn’t fair”) comes to us as a fil­i­gree of ash sto­ries. Walk­ing down the street past a cou­ple in con­ver­sa­tion, an over­heard mor­pheme, a mere glance at a wrongly but­toned rain­coat, sparks a nar­ra­tive in our imag­i­na­tion. Ask any ques­tion begin­ning with “why?” and the answer will surely be a story, or it will be embed­ded in a story. Or, at the very least, it will offer a tempt­ing thread for some story that you your­self will hold onto, embell­ish even, as you try to absorb the answer. We inter­po­late between such frag­ments. This is, for many of us, sim­ply the way we think. What about the “why ques­tions” in sci­ence, in logic, in math­e­mat­ics? We should acknowl­edge how they are often “what ques­tions” or “how ques­tions” in dis­guise. Or how they slide down into such ques­tions, as the ever-​​elusive, ever-​​illusory quest for an X that actu­ally causes a Y dis­solves. Some of the more sat­is­fy­ing answers to sci­en­tific “why” ques­tions involves deft rephras­ing. “Why is the sky blue?” is replaced by the ques­tion “what is the func­tion that describes scat­ter­ing ampli­tude as depen­dent on wave-​​length”?”

    math­e­mat­ics philosophy-​​of-​​mathematics sto­ry­telling prag­ma­tism theory-​​and-​​practice-​​sitting-​​in-​​a-​​tree what-​​is-​​it-​​good-​​for-​​hunh

Items of some interest:

Posted on by

These are my recent Pin​board​.in links:

  • Nicholas Rombes: Punk | berfrois

    “Most iron­i­cally, being based in the hope­lessly lost cul­tural void of Ann Arbor, a noto­ri­ous mecca for the last sur­viv­ing rem­nants of the pseudo-​​intellectual street peo­ple move­ment that said much and accom­plished little…”

    punk history-​​is-​​a-​​feature-​​not-​​a-​​bug cultural-​​dynamics ha-​​ha-​​only-​​semiserious

  • [1112.5309] POWERPLAY: Train­ing an Increas­ingly Gen­eral Prob­lem Solver by Con­tin­u­ally Search­ing for the Sim­plest Still Unsolv­able Problem

    An amus­ing col­lec­tion of what seem to be half-​​remembered ideas gleaned from his visit to the GPTP work­shop in Ann Arbor two years ago, pre­sented as his own inven­tions and with­out cita­tion or men­tion of the dozen peo­ple who actu­ally do this work. His keynote, as I remem­ber it, essen­tially revolved around him point­ing out how influ­en­tial his work should have been all along, if only we had both­ered to cite him as we should have done, because he thought up the core con­cepts of genetic pro­gram­ming well before any of us claimed we had. This is pretty much a camel’s-back straw for me. If there is a bet­ter argu­ment for com­pletely boy­cotting the cita­tion sys­tem and rely­ing on per­sonal asso­ci­a­tion and named schools rather than pub­li­ca­tion, I have not yet encoun­tered it. So remem­ber poor oppressed grad­u­ate and post­doc kids: when I cite your work by sim­ply nam­ing you per­son­ally, and not your advi­sor or your insti­tu­tion, and not even your pub­li­ca­tion or jour­nal but merely YOU PERSONALLY, it’s because you per­son­ally deserve the credit, not any of those other leeches. Got that?

    now-​​this-​​really-​​pisses-​​me-​​off-​​to-​​no-​​end

  • [1203.0856] Online Dis­crim­i­na­tive Dic­tio­nary Learn­ing for Image Clas­si­fi­ca­tion Based on Block-​​Coordinate Descent Method

    “Pre­vi­ous researches have demon­strated that the frame­work of dic­tio­nary learn­ing with sparse cod­ing, in which sig­nals are decom­posed as lin­ear com­bi­na­tions of a few atoms of a learned dic­tio­nary, is well adept to recon­struc­tion issues. This frame­work has also been used for dis­crim­i­na­tion tasks such as image clas­si­fi­ca­tion. To achieve bet­ter per­for­mances of clas­si­fi­ca­tion, experts develop sev­eral meth­ods to learn a dis­crim­i­na­tive dic­tio­nary in a super­vised man­ner. How­ever, another issue is that when the data become extremely large in scale, these meth­ods will be no longer effec­tive as they are all batch-​​oriented approaches. For this rea­son, we pro­pose a novel online algo­rithm for dis­crim­i­na­tive dic­tio­nary learn­ing, dubbed textbf{ODDL} in this paper. First, we intro­duce a lin­ear clas­si­fier into the con­ven­tional dic­tio­nary learn­ing for­mu­la­tion and derive a dis­crim­i­na­tive dic­tio­nary learn­ing prob­lem. Then, we exploit an online algo­rithm to solve the derived prob­lem. Unlike the most exist­ing approaches which update dic­tio­nary and clas­si­fier alter­nately via iter­a­tively solv­ing sub-​​problems, our approach directly explores them jointly. Mean­while, it can largely shorten the run­time for train­ing and is also par­tic­u­larly suit­able for large-​​scale clas­si­fi­ca­tion issues. To eval­u­ate the per­for­mance of the pro­posed ODDL approach in image recog­ni­tion, we con­duct some exper­i­ments on three well-​​known bench­marks, and the exper­i­men­tal results demon­strate ODDL is fairly promis­ing for image clas­si­fi­ca­tion tasks.”

    image-​​analysis image-​​segmentation algo­rithms nudge-​​targets

  • [1203.3271] The ther­mo­dy­nam­ics of prediction

    “A sys­tem respond­ing to a sto­chas­tic dri­ving sig­nal can be inter­preted as com­put­ing, by means of its dynam­ics, an (implicit) model of the envi­ron­men­tal vari­ables. The system’s state retains infor­ma­tion about past envi­ron­men­tal fluc­tu­a­tions, and a frac­tion of this infor­ma­tion is pre­dic­tive of future ones. The remain­ing non­pre­dic­tive infor­ma­tion reflects model com­plex­ity that does not improve pre­dic­tive power, and rep­re­sents the inef­fec­tive­ness of the model. We expose the fun­da­men­tal equiv­a­lence between this model inef­fi­ciency and ther­mo­dy­namic inef­fi­ciency, mea­sured by the energy dis­si­pated dur­ing the inter­ac­tion between sys­tem and envi­ron­ment. Our results hold arbi­trar­ily far from ther­mo­dy­namic equi­lib­rium and are applic­a­ble to a wide range of sys­tems, includ­ing bio­mol­e­c­u­lar machines. They high­light a pro­found con­nec­tion between the effec­tive use of infor­ma­tion and effi­cient ther­mo­dy­namic oper­a­tion: any sys­tem con­structed to keep mem­ory about its envi­ron­ment and to oper­ate ener­get­i­cally effi­ciently has to be predictive.”

    mod­el­ing philosophy-​​of-​​science information-​​theory physics ther­mo­dy­nam­ics talking-​​about-​​a-​​model-​​is-​​a-​​model pragmatism-it-ain’t

  • [1203.3434] On the Impact of Infor­ma­tion Tech­nolo­gies on Soci­ety: an His­tor­i­cal Per­spec­tive through the Game of Chess

    “The game of chess as always been viewed as an iconic rep­re­sen­ta­tion of intel­lec­tual prowess. Since the very begin­ning of com­puter sci­ence, the chal­lenge of being able to pro­gram a com­puter capa­ble of play­ing chess and beat­ing humans has been alive and used both as a mark to mea­sure hardware/​software pro­gresses and as an ongo­ing pro­gram­ming chal­lenge lead­ing to numer­ous dis­cov­er­ies. In the early days of com­puter sci­ence it was a topic for spe­cial­ists. But as com­put­ers were democ­ra­tized, and the strength of chess engines began to increase, chess play­ers started to appro­pri­ate to them­selves these new tools. We show how these inter­ac­tions between the world of chess and infor­ma­tion tech­nolo­gies have been her­ald of broader social impacts of infor­ma­tion tech­nolo­gies. The game of chess, and more broadly the world of chess (chess play­ers, lit­er­a­ture, com­puter soft­wares and web­sites ded­i­cated to chess, etc.), turns out to be a sur­pris­ingly and par­tic­u­larly sharp indi­ca­tor of the changes induced in our every­day life by the infor­ma­tion tech­nolo­gies. More­over, in the same way that chess is a mod­eliza­tion of war that cap­tures the raw fea­tures of strate­gic think­ing, chess world can be seen as small soci­ety mak­ing the study of the infor­ma­tion tech­nolo­gies impact eas­ier to ana­lyze and to grasp.”

    touch­stones his­tory algo­rithms history-​​of-​​science computer-​​science

  • Share Books | berfrois

    “Libraries are a recog­ni­tion that schol­ar­ship and cul­ture are more than the busi­ness of cre­at­ing and con­sum­ing. They are a human con­ver­sa­tion, and libraries pro­vide com­mon ground where that con­ver­sa­tion can take place and be remem­bered. By tak­ing aim at the right for the pub­lic to main­tain this con­ver­sa­tion and its mem­ory, pub­lish­ers have shown us what we have to lose. It’s time we resisted the out­sourc­ing of our com­mon her­itage by occu­py­ing the library.”

    Occupy libraries intellectual-​​property open-​​access public-​​policy activism

  • [1112.3307] Poly­tope Codes Against Adver­saries in Networks

    “Net­work cod­ing is stud­ied when an adver­sary con­trols a sub­set of nodes in the net­work of lim­ited quan­tity but unknown loca­tion. This prob­lem is shown to be more dif­fi­cult than when the adver­sary con­trols a given num­ber of edges in the net­work, in that lin­ear codes are insuf­fi­cient. To solve the node prob­lem, the class of Poly­tope Codes is intro­duced. Poly­tope Codes are con­stant com­po­si­tion codes oper­at­ing over bounded poly­topes in inte­ger vec­tor fields. The poly­tope struc­ture cre­ates addi­tional com­plex­ity, but it induces prop­er­ties on mar­ginal dis­tri­b­u­tions of code vec­tors so that validi­ties of code­words can be checked by inter­nal nodes of the net­work. It is shown that Poly­tope Codes achieve a cut-​​set bound for a class of pla­nar net­works. It is also shown that this cut-​​set bound is not always tight, and a tighter bound is given for an exam­ple network.”

    cryp­tog­ra­phy pri­vacy algo­rithms nudge-​​targets network-​​theory com­mu­ni­ca­tion

  • [1203.3353] Solv­ing Struc­ture with Sparse, Randomly-​​Oriented X-​​ray Data

    “Single-​​particle imag­ing exper­i­ments of bio­mol­e­cules at x-​​ray free-​​electron lasers (XFELs) require pro­cess­ing of hun­dreds of thou­sands (or more) of images that con­tain very few x-​​rays. Each low-​​flux image of the dif­frac­tion pat­tern is pro­duced by a sin­gle, ran­domly ori­ented par­ti­cle, such as a pro­tein. We demon­strate the fea­si­bil­ity of col­lect­ing data at these extremes, aver­ag­ing only 2.5 pho­tons per frame, where it seems doubt­ful there could be infor­ma­tion about the state of rota­tion, let alone the image con­trast. This is accom­plished with an expec­ta­tion max­i­miza­tion algo­rithm that processes the low-​​flux data in aggre­gate, and with­out any prior knowl­edge of the object or its ori­en­ta­tion. The ver­sa­til­ity of the method promises, more gen­er­ally, to rede­fine what mea­sure­ment sce­nar­ios can pro­vide use­ful sig­nal in the high-​​noise regime.”

    structural-​​biology image-​​analysis crys­tal­log­ra­phy algo­rithms inverse-​​problems nudge-​​targets sta­tis­tics

  • [1203.3203] An effi­cient algo­rithm for gen­er­at­ing AoA networks

    “The activ­i­ties, in project sched­ul­ing, can be rep­re­sented graph­i­cally in two dif­fer­ent ways, by either assign­ing the activ­i­ties to the nodes ‘AoN’ directed acyclic graph (dag) or to the arcs ‘AoA dag’. In this paper, a new algo­rithm is pro­posed for gen­er­at­ing, for a given project sched­ul­ing prob­lem, an Activity-​​on-​​Arc dag start­ing from the Activity-​​on-​​Node dag using the con­cepts of line graphs of graphs.”

    sched­ul­ing operations-​​research algo­rithms graph-​​theory

  • [1203.3341] A Com­par­i­son of Multi-​​Parametric Pro­gram­ming, Mixed-​​Integer Pro­gram­ming, Gra­di­ent Descent Based, and the Embed­ding Approach on Four Pub­lished Hybrid Opti­mal Con­trol Examples

    “…Com­mon mis­con­cep­tions regard­ing the embed­ding approach are addressed includ­ing whether or not it results in an aver­age value con­trol model (no), is nec­es­sary to “tweak” the algo­rithm to get bang-​​bang solu­tions (no), requires infi­nite switch­ing (no), has real-​​time capa­bil­ity (yes), or reduc­tion to a clas­si­cal non­lin­ear opti­miza­tion prob­lem (a desir­able yes).”

    control-​​theory operations-​​research algo­rithms numerical-​​methods philosophy-​​of-​​engineering design-​​patterns nudge-​​targets

  • [1203.3270] Extrac­tion of Facial Fea­ture Points Using Cumu­la­tive Histogram

    “This paper pro­poses a novel adap­tive algo­rithm to extract facial fea­ture points auto­mat­i­cally such as eye­brows cor­ners, eyes cor­ners, nos­trils, nose tip, and mouth cor­ners in frontal view faces, which is based on cumu­la­tive his­togram approach by vary­ing dif­fer­ent thresh­old val­ues. At first, the method adopts the Viola-​​Jones face detec­tor to detect the loca­tion of face and also crops the face region in an image. From the con­cept of the human face struc­ture, the six rel­e­vant regions such as right eye­brow, left eye­brow, right eye, left eye, nose, and mouth areas are cropped in a face image. Then the his­togram of each cropped rel­e­vant region is com­puted and its cumu­la­tive his­togram value is employed by vary­ing dif­fer­ent thresh­old val­ues to cre­ate a new fil­ter­ing image in an adap­tive way. The con­nected com­po­nent of inter­ested area for each rel­e­vant fil­ter­ing image is indi­cated our respec­tive fea­ture region. A sim­ple lin­ear search algo­rithm for eye­brows, eyes and mouth fil­ter­ing images and con­tour algo­rithm for nose fil­ter­ing image are applied to extract our desired cor­ner points auto­mat­i­cally. The method was tested on a large BioID frontal face data­base in dif­fer­ent illu­mi­na­tions, expres­sions and light­ing con­di­tions and the exper­i­men­tal results have achieved aver­age suc­cess rates of 95.27%.”

    image-​​segmentation image-​​analysis face-​​recognition algo­rithms nudge-​​targets

  • [1203.3284] Effi­cient Enu­mer­a­tion of the Directed Binary Per­fect Phy­lo­ge­nies from Incom­plete Data

    “We study a character-​​based phy­logeny recon­struc­tion prob­lem when an incom­plete set of data is given. More specif­i­cally, we con­sider the sit­u­a­tion under the directed per­fect phy­logeny assump­tion with binary char­ac­ters in which for some species the states of some char­ac­ters are miss­ing. Our main object is to give an effi­cient algo­rithm to enu­mer­ate (or list) all per­fect phy­lo­ge­nies that can be obtained when the miss­ing entries are com­pleted. While a sim­ple branch-​​and-​​bound algo­rithm (B&B) shows a the­o­ret­i­cally good per­for­mance, we pro­pose another approach based on a zero-​​suppressed binary deci­sion dia­gram (ZDD). Exper­i­men­tal results on ran­domly gen­er­ated data exhibit that the ZDD approach out­per­forms B&B. We also prove that count­ing the num­ber of phy­lo­ge­netic trees con­sis­tent with a given data is #P-​​complete, thus pro­vid­ing an evi­dence that an effi­cient ran­dom sam­pling seems hard.”

    phy­lo­ge­net­ics inverse-​​problems genet­ics algo­rithms sta­tis­tics nudge-​​targets

  • [1203.0879] Design­ing and using prior knowl­edge for phase retrieval

    “In this work we develop an algo­rithm for sig­nal recon­struc­tion from the mag­ni­tude of its Fourier trans­form in a sit­u­a­tion where some (non-​​zero) parts of the sought sig­nal are known. Although our method does not assume that the known part com­prises the bound­ary of the sought sig­nal, this is often the case in microscopy: a spec­i­men is placed inside a known mask, which can be thought of as a known light source that sur­rounds the unknown sig­nal. There­fore, in the past, sev­eral algo­rithms were sug­gested that solve the phase retrieval prob­lem assum­ing known bound­ary val­ues. Unlike our method, these meth­ods do rely on the fact that the known part is on the bound­ary. Besides the recon­struc­tion method we give an expla­na­tion of the phe­nom­ena observed in pre­vi­ous work: the recon­struc­tion is much faster when there is more energy con­cen­trated in the known part. Quite sur­pris­ingly, this can be explained using our pre­vi­ous results on phase retrieval with approx­i­mately known Fourier phase.”

    image-​​analysis image-​​processing learn­ing inverse-​​problems algo­rithms nudge-​​targets

  • [1203.3415] A New Approach to Count Pat­tern Motifs Using Com­bi­na­to­r­ial Techniches

    “We pro­posed two new exact algo­rithms to detect net­work motifs of size 3 and 4. Con­sid­er­ing that motifs need to count the iso­mor­phic pat­terns in the orig­i­nal graph $G(V,E)$ and in a set of ran­dom­ized graphs, the fol­low­ing com­plex­i­ties con­cern about count iso­mor­phic pat­terns in a sin­gle graph. Let $m=|E|$ and let $a(G)$ be the arboric­ity of $G$. Assume $|E|geq|V|$. We describe a $O(a(G)m)$ time com­plex­ity algo­rithm to count iso­mor­phic pat­terns of size 3. The com­plex­ity is a $O({msqrt{m}})$ in the worst graph. The sec­ond algo­rithm is a $O(m^2)$ com­plex­ity algo­rithm to count iso­mor­phic pat­terns of size 4. The final result was expres­sive faster when com­pared with other imple­mented algorithms.”

    network-​​theory graph-​​theory algo­rithms nudge-​​targets

Items of some interest:

Posted on by

These are my recent Pin​board​.in links:

Items of some interest:

Posted on by

These are my recent Pin​board​.in links:

  • Wel­come to the Group Pat­tern Lan­guage Project | Group Works

    “This deck of 91 full-​​colour cards names what skilled facil­i­ta­tors and other par­tic­i­pants do to make things work.  The con­tent is more spe­cific than val­ues and less spe­cific than tips and tech­niques, cut­ting across exist­ing method­olo­gies with a designer’s eye to cap­ture the pat­terns that repeat.  The deck can be used to plan sess­sions, reflect on and debrief them, pro­vide guid­ance, and share respon­si­bil­ity for mak­ing the process go well.  It has the poten­tial to pro­vide a com­mon ref­er­ence point for prac­ti­tion­ers, and serve as a frame­work and learn­ing tool for those study­ing the field. ”

    via:bkerr col­lab­o­ra­tion design-​​patterns tools social-​​dynamics

  • [1202.0001] Vector-​​based model of elas­tic bonds for DEM sim­u­la­tion of solids

    “A new model for com­puter sim­u­la­tion of solids, com­posed of bonded par­ti­cles, is pro­posed. Vec­tors rigidly con­nected with par­ti­cles are used for descrip­tion of defor­ma­tion of a sin­gle bond. The expres­sion for poten­tial energy of the bond and cor­re­spond­ing expres­sions for forces and moments are pro­posed. For­mu­las, con­nect­ing para­me­ters of the model with lon­gi­tu­di­nal, shear, bend­ing and tor­sional stiff­nesses of the bond, are derived. It is shown that the model allows to describe any val­ues of the bond stiff­nesses exactly. Two dif­fer­ent cal­i­bra­tion pro­ce­dures depend­ing on bond length/​thickness ratio are pro­posed. It is shown that para­me­ters of model can be cho­sen so that under small defor­ma­tions the bond is equiv­a­lent to either Bernoulli-​​Euler or Tim­o­shenko rod or short cylin­der con­nect­ing par­ti­cles. Sim­ple expres­sions, con­nect­ing para­me­ters of V-​​model with geo­met­ri­cal and mechan­i­cal char­ac­ter­is­tics of the bond, are derived. Com­puter sim­u­la­tion of dynam­i­cal buck­ling of the straight dis­crete rod and dis­crete half-​​spherical shell is car­ried out.”

    mod­el­ing mechanical-​​systems materials-​​science computational-​​methods algo­rithms nudge-​​targets

  • [1202.0253] High-​​speed Flight in an Ergodic Forest

    “Inspired by birds fly­ing through clut­tered envi­ron­ments such as dense forests, this paper stud­ies the the­o­ret­i­cal foun­da­tions of a novel motion plan­ning prob­lem: high-​​speed nav­i­ga­tion through a randomly-​​generated obsta­cle field when only the sta­tis­tics of the obsta­cle gen­er­at­ing process are known a pri­ori. Resem­bling a pla­nar for­est envi­ron­ment, the obsta­cle gen­er­at­ing process is assumed to deter­mine the loca­tions and sizes of disk-​​shaped obsta­cles. When this process is ergodic, and under mild tech­ni­cal con­di­tions on the dynam­ics of the bird, it is shown that the exis­tence of an infi­nite collision-​​free tra­jec­tory through the for­est exhibits a phase tran­si­tion. On one hand, if the bird flies faster than a cer­tain crit­i­cal speed, then, with prob­a­bil­ity one, there is no infi­nite collision-​​free tra­jec­tory, i.e., the bird will even­tu­ally col­lide with some tree, almost surely, regard­less of the plan­ning algo­rithm gov­ern­ing the bird’s motion. On the other hand, if the bird flies slower than this crit­i­cal speed, then there exists at least one infi­nite collision-​​free tra­jec­tory, almost surely. Lower and upper bounds on the crit­i­cal speed are derived for the spe­cial case of a homo­ge­neous Pois­son for­est con­sid­er­ing a sim­ple model for the bird’s dynam­ics. For the same case, an equiv­a­lent per­co­la­tion model is pro­vided. Using this model, the phase dia­gram is approx­i­mated in Monte-​​Carlo sim­u­la­tions. This paper also estab­lishes novel con­nec­tions between robot motion plan­ning and sta­tis­ti­cal physics through ergodic the­ory and per­co­la­tion the­ory, which may be of inde­pen­dent interest.”

    robot­ics plan­ning algo­rithms nudge-​​targets

  • [1202.0077] An Inter­act­ing Par­ti­cle Model for Clus­ter­ing Euclid­ean Datasets

    “In this paper we pro­pose a method based on inter­act­ing par­ti­cle physics, devised for clus­ter­ing Euclid­ean datasets with­out ini­tial con­straints or con­di­tions. We model any dataset as an inter­act­ing par­ti­cle sys­tem, whose ele­ments cor­re­spond to par­ti­cles that inter­act through a sim­pli­fied ver­sion of Lennard-​​Jones poten­tials. In so doing, mutual attrac­tive inter­ac­tions allow to iden­tify groups of prox­i­mal par­ti­cles. The main out­come of this mod­el­ing task is an adja­cency matrix, taken as input by a com­mu­nity detec­tion algo­rithm aimed to iden­tify dif­fer­ent par­ti­tions. The under­ly­ing con­jec­ture is that, using a mul­tires­o­lu­tion analy­sis, the adopted model allows to find the right num­ber of clus­ters for any given dataset. Exper­i­men­tal results, per­formed in com­par­i­son with a clas­si­cal clus­ter­ing algo­rithm, con­firm this assumption.”

    clus­ter­ing data-​​analysis algo­rithms nudge-​​targets distributed-​​processing

  • [1201.6583] Empow­er­ment for Con­tin­u­ous Agent-​​Environment Systems

    “This paper devel­ops gen­er­al­iza­tions of empow­er­ment to con­tin­u­ous states. Empow­er­ment is a recently intro­duced information-​​theoretic quan­tity moti­vated by hypothe­ses about the effi­ciency of the sen­so­ri­mo­tor loop in bio­log­i­cal organ­isms, but also from con­sid­er­a­tions stem­ming from curiosity-​​driven learn­ing. Empowe­mer­ment mea­sures, for agent-​​environment sys­tems with sto­chas­tic tran­si­tions, how much influ­ence an agent has on its envi­ron­ment, but only that influ­ence that can be sensed by the agent sen­sors. It is an information-​​theoretic gen­er­al­iza­tion of joint con­trol­la­bil­ity (influ­ence on envi­ron­ment) and observ­abil­ity (mea­sure­ment by sen­sors) of the envi­ron­ment by the agent, both con­trol­la­bil­ity and observ­abil­ity being usu­ally defined in con­trol the­ory as the dimen­sion­al­ity of the control/​observation spaces.…”

    agent-​​based emergent-​​design robot­ics engineering-​​design machine-​​learning empow­er­ment nudge

  • [1201.6655] Learn­ing Per­for­mance of Pre­dic­tion Mar­kets with Kelly Bettors

    “In eval­u­at­ing pre­dic­tion mar­kets (and other crowd-​​prediction mech­a­nisms), inves­ti­ga­tors have repeat­edly observed a so-​​called “wis­dom of crowds” effect, which roughly says that the aver­age of par­tic­i­pants per­forms much bet­ter than the aver­age par­tic­i­pant. The mar­ket price—an aver­age or at least aggre­gate of traders’ beliefs—offers a bet­ter esti­mate than most any indi­vid­ual trader’s opin­ion. In this paper, we ask a stronger ques­tion: how does the mar­ket price com­pare to the best trader’s belief, not just the aver­age trader. We mea­sure the market’s worst-​​case log regret, a notion com­mon in machine learn­ing the­ory. To arrive at a mean­ing­ful answer, we need to assume some­thing about how traders behave. We sup­pose that every trader opti­mizes accord­ing to the Kelly cri­te­ria, a strat­egy that prov­ably max­i­mizes the com­pound growth of wealth over an (infi­nite) sequence of mar­ket inter­ac­tions. We show sev­eral consequences.…”

    pre­dic­tion performance-​​measure agent-​​based sim­u­la­tion nudge-​​targets wisdom-​​of-​​crowds

  • Curat­ing the kraken « Pub­lic Historian

    ‘This is why “curate” is still a word to con­jure by in our cul­ture.  It still promises trans­for­ma­tive power.’

    muse­ol­ogy prag­mat­ics nam­ing engineering-​​of-​​philosophy

  • [1201.5780] Full and Half Gilbert Tes­sel­la­tions with Rec­tan­gu­lar Cells

    “We inves­ti­gate the ray-​​length dis­tri­b­u­tions for two dif­fer­ent rec­tan­gu­lar ver­sions of Gilbert’s tes­sel­la­tion. In the full rec­tan­gu­lar ver­sion, lines extend either hor­i­zon­tally (with east– and west-​​growing rays) or ver­ti­cally (north– and south-​​growing rays) from seed points which form a Pois­son point process, each ray stop­ping when another ray is met. In the half rec­tan­gu­lar ver­sion, east and south grow­ing rays do not inter­act with west and north rays. For the half rec­tan­gu­lar tes­sel­la­tion we com­pute ana­lyt­i­cally, via recur­sion, a series expan­sion for the ray-​​length dis­tri­b­u­tion, whilst for the full rec­tan­gu­lar ver­sion we develop an accu­rate sim­u­la­tion tech­nique, based in part on the stopping-​​set the­ory of Zuyev, to accom­plish the same. We demon­strate the remark­able fact that plots of the two dis­tri­b­u­tions appear to be iden­ti­cal when the inten­sity of seeds in the half model is twice that in the full model. Our paper explores this coin­ci­dence mind­ful of the fact that, for one model, our results are from a sim­u­la­tion (with inher­ent sam­pling error).…”

    geom­e­try tiling algo­rithms generative-​​art sim­u­la­tion emer­gence interesting-​​problem

Items of some interest…

Posted on by

These are my recent Pin​board​.in links:

  • [1107.0674] “Mem­ory foam” approach to unsu­per­vised learning

    “We pro­pose an alter­na­tive approach to con­struct an arti­fi­cial learn­ing sys­tem, which nat­u­rally learns in an unsu­per­vised man­ner. Its math­e­mat­i­cal pro­to­type is a dynam­i­cal sys­tem, which auto­mat­i­cally shapes its vec­tor field in response to the input sig­nal. The vec­tor field con­verges to a gra­di­ent of a multi-​​dimensional prob­a­bil­ity den­sity dis­tri­b­u­tion of the input process, taken with neg­a­tive sign. The most prob­a­ble pat­terns are rep­re­sented by the sta­ble fixed points, whose basins of attrac­tion are formed auto­mat­i­cally. The per­for­mance of this sys­tem is illus­trated with musi­cal signals.”

    machine-​​learning clas­si­fi­ca­tion learning-​​from-​​data algo­rithms nudge-​​targets

  • [1107.0550] 3D Ter­res­trial LiDAR data clas­si­fi­ca­tion of com­plex nat­ural scenes using a multi-​​scale dimen­sion­al­ity cri­te­rion: appli­ca­tions in geomorphology

    “3D point clouds of nat­ural envi­ron­ments rel­e­vant to geo­mor­phol­ogy prob­lems (rivers, cliffs…) often require to clas­sify the data into ele­men­tary rel­e­vant classes. A typ­i­cal exam­ple is the sep­a­ra­tion of ripar­ian veg­e­ta­tion from soil in flu­vial envi­ron­ments, the dis­tinc­tion between fresh sur­faces and rock­fall in cliff envi­ron­ments, or more gen­er­ally the clas­si­fi­ca­tion of sur­faces accord­ing to their mor­phol­ogy (rip­ples, grain size…). Nat­ural sur­faces are very het­ero­ge­neous and their dis­tinc­tive prop­er­ties are sel­dom defined at a unique scale. We have thus defined a multi-​​scale mea­sure of the point cloud dimen­sion­al­ity around each point. The dimen­sion­al­ity char­ac­ter­izes the local 3D orga­ni­za­tion of the point cloud and varies from being 1D (points set along a line) to really tak­ing all 3D vol­ume, at each scale. We present the tech­nique and illus­trate its effi­ciency in sep­a­rat­ing ripar­ian veg­e­ta­tion from ground and clas­si­fy­ing a moun­tain stream in veg­e­ta­tion, rock, gravel and water sur­face. The supe­ri­or­ity of the multi-​​scale analy­sis in enhanc­ing class sep­a­ra­bil­ity and spa­tial res­o­lu­tion of the clas­si­fi­ca­tion is also demon­strated. Large scenes can be clas­si­fied on a com­mod­ity lap­top in a rea­son­able time. The tech­nique is robust to miss­ing data and espe­cially shadow zones. The clas­si­fi­ca­tion is fast and accu­rate and can account for some degree of intra-​​class mor­pho­log­i­cal vari­abil­ity such as dif­fer­ent veg­e­ta­tion types. A prob­a­bilis­tic con­fi­dence in the clas­si­fi­ca­tion result is given at each point allow­ing the user to remove the points for which the clas­si­fi­ca­tion is uncer­tain. The process can be both fully auto­mated but also fully cus­tomized by the user includ­ing a graph­i­cal def­i­n­i­tion of the clas­si­fiers if so desired. Although devel­oped for fully 3D data, the method can be read­ily applied to 2.5D air­borne LiDAR data.”

    image-​​analysis image-​​segmentation learning-​​from-​​data clas­si­fi­ca­tion nudge-​​targets

  • [1105.6001] A Call to Arms: Revis­it­ing Data­base Design

    “Good data­base design is cru­cial to obtain a sound, con­sis­tent data­base, and — in turn — good data­base design method­olo­gies are the best way to achieve the right design. These method­olo­gies are taught to most Com­puter Sci­ence under­grad­u­ates, as part of any Intro­duc­tion to Data­base class. They can be con­sid­ered part of the “canon”, and indeed, the over­all approach to data­base design has been unchanged for years. More­over, none of the major data­base research assess­ments iden­tify data­base design as a strate­gic research direc­tion. Should we con­clude that data­base design is a solved prob­lem? Our the­sis is that data­base design remains a crit­i­cal unsolved prob­lem. Hence, it should be the sub­ject of more research. Our start­ing point is the obser­va­tion that tra­di­tional data­base design is not used in prac­tice — and if it were used it would result in designs that are not well adapted to cur­rent envi­ron­ments. In short, data­base design has failed to keep up with the times. In this paper, we put forth argu­ments to sup­port our view­point, ana­lyze the root causes of this sit­u­a­tion and sug­gest some avenues of research.”

    data­base ontol­ogy software-​​development computer-​​science design-​​patterns