Items of some interest:

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

  • [1206.3340] Extrac­tion of Deep Phy­lo­ge­netic Sig­nal and Improved Res­o­lu­tion of Evo­lu­tion­ary Events within the recA/​RAD51 Phylogeny

    “The recA/​RAD51 gene fam­ily encodes a diverse set of recom­bi­nase pro­teins that effect homol­o­gous recom­bi­na­tion, DNA-​​repair, and genome sta­bil­ity. The recA gene fam­ily is expressed in almost all species of Eubac­te­ria, Archaea, and Eukary­otes, and even in some viruses. To date, efforts to resolve the deep evo­lu­tion­ary ori­gins of this ancient pro­tein fam­ily have been hin­dered, in part, by the high sequence diver­gence between fam­i­lies (i.e. ~30% iden­tity between par­al­o­gous groups). Through (i) large taxon sam­pling, (ii) the use of a phy­lo­ge­netic algo­rithm designed for mea­sur­ing highly diver­gent par­alogs, and (iii) novel Evo­lu­tion­ary Spa­tial Dynam­ics sim­u­la­tion and ana­lyt­i­cal tools, we obtained a robust, par­si­mo­nious and more refined phy­lo­ge­netic his­tory of the recA/​RAD51 super­fam­ily. Taken together, our model for the evo­lu­tion of recA/​RAD51 fam­ily pro­vides a bet­ter under­stand­ing of ancient ori­gin of recA pro­teins and mul­ti­ple events lead­ing to the diver­si­fi­ca­tion of recA homologs in eukary­otes, includ­ing the dis­cov­ery of addi­tional RAD51 sub-​​families.”

    cladis­tics algo­rithms visu­al­iza­tion deep-​​time sta­tis­tics
  • [1204.6547] Gen­er­at­ing self-​​organizing col­lec­tive behav­ior using sep­a­ra­tion dynam­ics from exper­i­men­tal data

    “Math­e­mat­i­cal mod­els for sys­tems of inter­act­ing agents using sim­ple local rules have been pro­posed and shown to exhibit emer­gent swarm­ing behav­ior. Most of these mod­els are con­structed by intu­ition or man­ual obser­va­tions of real phe­nom­ena, and later tuned or ver­i­fied to sim­u­late desired dynam­ics. In con­trast to this approach, we pro­pose using a model that attempts to fol­low an aver­aged rule of the essen­tial distance-​​dependent col­lec­tive behav­ior of real pigeon flocks, which was abstracted from exper­i­men­tal data. By using a sim­ple model to fol­low the behav­ioral ten­den­cies of real data, we show that our model can exhibit emer­gent self-​​organizing dynam­ics such as flock­ing, pat­tern for­ma­tion, and counter-​​rotating vor­tices. The range of behav­iors observed in our sim­u­la­tions are richer than the stan­dard mod­els of col­lec­tive dynam­ics, and should thereby give poten­tial for new mod­els of com­plex behavior.”

    agent-​​based swarms boids algo­rithms emergent-​​design
  • [1206.3555] A Dynamic Pro­gram­ming Algo­rithm for Infer­ence in Recur­sive Prob­a­bilis­tic Programs

    “We describe a dynamic pro­gram­ming algo­rithm for com­put­ing the mar­ginal dis­tri­b­u­tion of dis­crete prob­a­bilis­tic pro­grams. This algo­rithm takes a func­tional inter­preter for an arbi­trary prob­a­bilis­tic pro­gram­ming lan­guage and turns it into an effi­cient mar­gin­al­izer. Because direct caching of sub-​​distributions is impos­si­ble in the pres­ence of recur­sion, we build a graph of depen­den­cies between sub-​​distributions. This fac­tored sum-​​product net­work makes (poten­tially cyclic) depen­den­cies between sub­prob­lems explicit, and cor­re­sponds to a sys­tem of equa­tions for the mar­ginal dis­tri­b­u­tion. We solve these equa­tions by fixed-​​point iter­a­tion in topo­log­i­cal order. We illus­trate this algo­rithm on exam­ples used in teach­ing prob­a­bilis­tic mod­els, com­pu­ta­tional cog­ni­tive sci­ence research, and game theory.”

    recur­sion stochastic-​​programming sim­u­la­tion nudge
  • Share­able: Hack­ing Home: Col­iv­ing Rein­vents the Com­mune for a Net­worked Age

    ‘It was more than just a lux­ury home full of bril­liant young minds. Dubbed “an inten­tional com­mu­nity”, The Rain­bow Man­sion was an exper­i­ment in a new type of cohab­i­ta­tion. The house began host­ing hackathons and salons in its library, invit­ing Sil­i­con Valley’s best and bright­est to par­tic­i­pate. “Right away it set itself in motion,” Schin­gler says. “It had this sort of acci­den­tal mys­tique about it.”’

    cohous­ing col­lab­o­ra­tion nerd-​​culture
  • [1206.3369] A Suc­ces­sive Approx­i­ma­tion Algo­rithm for Com­put­ing the Divi­sor Sum­ma­tory Function

    “An algo­rithm is pre­sented to com­pute iso­lated val­ues of the divi­sor sum­ma­tory func­tion in O(n^(1/3)) time and O (log n) space. The algo­rithm is ele­men­tary and uses a geo­met­ric approach of suc­ces­sive approx­i­ma­tion com­bined with coor­di­nate transformation.”

    algo­rithms computational-​​geometry nudge-​​targets
  • [1204.3650] Evo­lu­tion­ary Meta­dy­nam­ics: a Novel Method to Pre­dict Crys­tal Structures

    “A novel method for crys­tal struc­ture pre­dic­tion, based on meta­dy­nam­ics and evo­lu­tion­ary algo­rithms, is pre­sented here. This tech­nique can be used to pro­duce effi­ciently both the ground state and metastable states eas­ily reach­able from a rea­son­able ini­tial struc­ture. We use the cell shape as col­lec­tive vari­able and evo­lu­tion­ary vari­a­tion oper­a­tors devel­oped in the con­text of the USPEX method [Oganov, Glass, textit{J. Chem. Phys.}, 2006, textbf{124}, 244704; Lyakhov textit{et al., Comp. Phys. Comm.}, 2010, textbf{181}, 1623; Oganov textit{et al., Acc. Chem. Res.}, 2011, textbf{44}, 227] to equi­li­brate the sys­tem as a func­tion of the col­lec­tive vari­ables. We illus­trate how this approach helps one to find sta­ble and metastable states for Al$_2$SiO$_5$, SiO$_2$, MgSiO$_3$, and car­bon. Apart from pre­dict­ing crys­tal struc­tures, the new method can also pro­vide insight into mech­a­nisms of phase transitions.”

    evolutionary-​​algorithms search-​​algorithms physics nudge-​​targets condensed-​​matter

Items of some interest:

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

  • What if Inter­ac­tiv­ity is the New Pas­siv­ity? Jonathan Sterne /​ McGill Uni­ver­sity | Flow

    “What if all the bad things that media crit­ics have been said about pas­siv­ity for the past cen­tury or two are now equally applic­a­ble to all the demands to inter­act, to par­tic­i­pate? What if inter­ac­tiv­ity is now one of the cen­tral hinges through which power works? In many moments today, the most com­pli­ant ges­ture we can make is to con­sent to inter­act on the terms pre­sented to us by our soft­ware and machines. This pull is espe­cially strong in those com­mer­cial plat­forms that cel­e­brate their own dif­fer­ence from the so-​​called pas­sive media of pre­vi­ous decades, and in the process mon­e­tize their users’ par­tic­i­pa­tion either directly or indi­rectly. What if—from time to time—we chose not to iden­tify with the inter­ac­tive promise of new media plat­forms or for that mat­ter new media art? What if, when the new media savants lam­bast so-​​called old media audi­ences as denizens of pas­siv­ity and ide­ol­ogy, we say, “yes, that’s me”?”

    a-​​bit-​​too-​​theoryish cultural-​​norms ingroup-​​outgroup new-​​media
  • How Can Her­bert Spencer’s 1892 Revi­sions to his Social Sta­t­ics Help Us Under­stand Con­ser­v­a­tive Oppo­si­tion to the Indi­vid­ual Man­date? | Rortybomb

    “But I think it’s clear what his real objec­tion was: uni­ver­sal suf­frage has the poten­tial to advance social­is­tic causes, inter­fer­ing with his laissez-​​faire project. From his auto­bi­og­ra­phy: “Another exten­sion of the fran­chise since made…will inevitably be fol­lowed by a still more rapid growth of social­is­tic leg­is­la­tion.” When he real­ized women’s equal­ity could poten­tially inter­fere with laissez-​​faire eco­nom­ics, it was time for women’s equal­ity to get cut from his over­all the­ory of a bet­ter world. He would rather muti­late his intel­lec­tual project instead of allow­ing his ene­mies to con­tinue to build their gov­er­nance project.”

    Herbert-​​Spencer laissez-​​faire cor­po­ratism cap­i­tal­ism pol­i­tics con­ser­vatism via:cshalizi
  • BloJJ — About con­fer­ence poster design and defense:

    “My approach is dif­fer­ent. Poster pre­sen­ta­tion, like con­fer­ence pre­sen­ta­tion, belongs more to the area of dra­matic arts than to mar­ket­ing. It is information/​entertainment, and that is the main thing you have to bear in mind when prepar­ing for the ses­sion. Plus, while at a con­fer­ence you have the full atten­tion of your audi­ence (shared, of course, with email, Face­book, plus the 10% that are sim­ply speak­ing) in a poster ses­sion you have to first attract the atten­tion of the peo­ple wan­der­ing around a hall shared with other 20 to 100 posters, then keep them there for the dura­tion of the spiel and while you start a new one, and then, of course, con­vey the infor­ma­tion you want to share with your poster. ”

    advice academic-​​culture meet­ing poster-​​presentaitons skills
  • Economist’s View: The 999

    “Some Indi­vid­u­als of our Coun­try­men, by the Smiles of Prov­i­dence or some other Means, are enabled to roll in their four–wheel’d Car­riages, and can sup­port the Expence of good Houses, rich Fur­ni­ture, and Lux­u­ri­ous Liv­ing. But, is it equi­table that 99, or rather 999 should suf­fer for the Extrav­a­gance or Grandeur of one? Espe­cially when it is consider’d, that Men fre­quently owe their Wealth to the Impov­er­ish­ment of their Neighbours.”

    it-​​was-​​ever-​​thus
  • Ris­ingTide­Har­bor: Matt Barcomb’s Blog on Lean Agile Busi­ness Soft­ware Devel­op­ment: Stop B*tching About Local Optimizations

    “In fact, one approach is to inten­tion­ally over opti­mize a local opti­miza­tion. This will often make appar­ent to man­age­ment (or even to you) where the true bot­tle neck in the sys­tem is. We shouldn’t worry so much about doing the wrong things righter, but we should be aware that that may be the case and always work to be doing the right things. In the end, show­ing improve­ment and build­ing momen­tum can lead to excit­ing changes. In fair­ness, it can also come crash­ing to the ground if the right kinds of changes aren’t made at some point, but this should not deter any­one who thinks some­thing can be made bet­ter from try­ing to do so and it cer­tainly should not be a rea­son to do nothing!”

    change cultural-​​engineering organizational-​​behavior local-​​optimization
  • Geof­frey Chaucer Hath a Blog: A Long Tyme Agoon in a Shire Far Away

    “…A WHINY YOUTHE cam nexte, barl­eye a man, With yelwe haire, tunique, and farmeres tan. But aqua­cul­ture litel did he love, He wolde been a pilot al above And bulls­eye oump-​​rattes yn a nim­ble craft.…”

    amus­ing
  • knitr: Ele­gant, flex­i­ble and fast dynamic report gen­er­a­tion with R | knitr

    “The knitr pack­age was designed to be a trans­par­ent engine for dynamic report gen­er­a­tion with R, solve some long-​​standing prob­lems in Sweave, and com­bine fea­tures in other add-​​on pack­ages into one pack­age (knitr ≈ Sweave + cacheSweave + pgf­Sweave + weaver + R2HTML::RweaveHTML + highlight::HighlightWeaveLatex + 0.2 * brew + 0.1 * SweaveListingUtils + more).”

    R-​​language LaTeX type­set­ting dynamic-​​documents writ­ing tools

  • nudge-​​targets mathematical-​​recreations
  • Cere­bral Mastication

    “There’s a charm­ing lit­tle brain teaser that’s going around the Inter­webs. It’s got var­i­ous forms, but they all look some­thing like this:…”

    nudge-​​targets mathematical-​​recreations
  • Tanya Khovanova’s Math Blog » Blog Archive » Inter­lock­ing Polyominoes

    “A set of poly­omi­noes is inter­locked if no sub­set can be moved far away from the rest. It was known that poly­omi­noes that are built from four or fewer squares do not inter­lock. The project of Dhawan and his men­tor was to inves­ti­gate the inter­locked­ness of larger poly­omi­noes. And they totally deliv­ered. They quickly proved that you can inter­lock poly­omi­noes with eight or more squares. Then they proved that pen­tomi­noes can’t inter­lock. This left them with a gray area: what hap­pens with poly­omi­noes with six or seven squares? After draw­ing many beau­ti­ful pic­tures, they finally found the struc­ture pre­sented in our accom­pa­ny­ing image. The sys­tem con­sists of 12 hex­omi­noes and 5 pen­tomi­noes, and it is rigid. You can­not move a thing. That means that hex­omi­noes can be inter­locked and thus the gray area was resolved.”

    poly­omi­noes mathematical-​​recreations nudge-​​targets
  • Pool based evo­lu­tion­ary algo­rithm pre­sented in EvoStar 2012 « GeNeura Team

    “This is the first inter­na­tion­ally pub­lished paper (it was pre­vi­ously pub­lished in a Span­ish con­fer­ence of a series that deals with a sys­tem, intended for vol­un­teer com­put­ing, that uses a pool for imple­ment­ing dis­trib­uted evo­lu­tion­ary algo­rithms. The basic idea is that the pop­u­la­tion resides in a pool (imple­mented using CouchDB), with clients pulling indi­vid­u­als from the pool, doing stuff on them, and putting them back in the pool. The algo­rithm uses, as much as pos­si­ble, CouchDB fea­tures (such as revi­sions and views) to achieve good per­for­mance. All the code (for this and, right now, for the next papers) is avail­able as open-​​source code.”

    distributed-​​processing evolutionary-​​algorithms CouchDB nudge
  • What Amazon’s ebook strat­egy means — Charlie’s Diary

    “If the major pub­lish­ers switch to sell­ing ebooks with­out DRM, then they can enable cus­tomers to buy books from a vari­ety of out­lets and move away from the walled gar­den of the Kin­dle store. They see DRM as a defense against piracy, but piracy is a much less imme­di­ate threat than a gigan­tic multi­na­tional with rev­enue of $48 Bil­lion in 2011 (more than the entire global pub­lish­ing indus­try) that has expressed its inten­tion to “dis­rupt” them, and whose chief exec­u­tive said recently “even well-​​meaning gate­keep­ers slow inno­va­tion” (where “inno­va­tion” is code-​​speak for “oppor­tu­ni­ties for me to turn a profit”). And so they will deep-​​six their exist­ing com­mit­ment to DRM and use the terms of the DoJ-​​imposed set­tle­ment to wig­gle out of the most-​​favoured-​​nation terms imposed by Ama­zon, in order to sell their wares as widely as pos­si­ble. If they don’t, they’re doomed. And all of us who like to read (or write) fic­tion get to live in the Ama­zon com­pany town.”

    monopoly-​​and-​​monpsony-​​sittin-​​in-​​a-​​tree Ama­zon eBooks disintermediation-​​in-​​action cor­po­ratism redis­in­ter­me­di­a­tion

Items of some interest:

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

  • [1201.5440] Self-​​assembly of anisotropic soft par­ti­cles in two dimensions

    “The self assem­bly of core-​​corona discs inter­act­ing via anisotropic poten­tials is inves­ti­gated using Monte Carlo com­puter sim­u­la­tions. A min­i­mal inter­ac­tion poten­tial that incor­po­rates anisotropy in a sim­ple way is intro­duced. It con­sists in a core-​​corona archi­tec­ture in which the cen­ter of the core is shifted with respect to the cen­ter of the corona. Anisotropy can thus be tuned by pro­gres­sively shift­ing the posi­tion of the core. Despite its sim­plic­ity, the sys­tem self orga­nize in a rich vari­ety of struc­tures includ­ing stripes, tri­an­gu­lar and rec­tan­gu­lar lat­tices, and unusual plas­tic crys­tals. Our results indi­cate that the amount of anisotropy does not alter the lat­tice spac­ing and only influ­ences the type of clus­ter­ing (stripes, micells, etc.) of the indi­vid­ual particles.”

    self-​​assembly biologically-​​inspired sim­u­la­tion pattern-​​formation condensed-​​matter
  • [1201.5477] Entropy-​​growth-​​based model of emo­tion­ally charged online dialogues

    “We ana­lyze emo­tion­ally anno­tated mas­sive data from IRC (Inter­net Relay Chat) and model the dia­logues between its par­tic­i­pants by assum­ing that the dri­ving force for the dis­cus­sion is the entropy growth of emo­tional prob­a­bil­ity dis­tri­b­u­tion. This process is claimed to be cor­re­lated to the emer­gence of the power-​​law dis­tri­b­u­tion of the dis­cus­sion lengths observed in the dia­logues. We per­form numer­i­cal sim­u­la­tions based on the noticed phe­nom­e­non obtain­ing a good agree­ment with the real data. Finally, we pro­pose a method to arti­fi­cially pro­long the dura­tion of the dis­cus­sion that relies on the entropy of emo­tional prob­a­bil­ity distribution.”

    oh-​​look-​​power-​​laws flame-​​wars social-​​dynamics com­plex­ol­ogy cultural-​​dynamics
  • [1201.4955] Coor­di­na­tion, Dif­fer­en­ti­a­tion and Fair­ness in a pop­u­la­tion of coop­er­at­ing agents

    “In a recent paper, we ana­lyzed the self-​​assembly of a com­plex coop­er­a­tion net­work. The net­work was shown to approach a state, where every agent invests the same amount of resources. Nev­er­the­less, highly-​​connected agents arise that extract extra-​​ordinarily high pay­offs while con­tribut­ing com­pa­ra­bly lit­tle to any of their coop­er­a­tions. Here, we inves­ti­gate a vari­ant of the model, in which highly-​​connected agents have access to addi­tional resources. We study ana­lyt­i­cally and numer­i­cally whether these resources are invested in exist­ing col­lab­o­ra­tions, lead­ing to a fairer load dis­tri­b­u­tion, or in estab­lish­ing new col­lab­o­ra­tions, lead­ing to an even less fair dis­tri­b­u­tion of loads and payoffs.”

    col­lab­o­ra­tion social-​​capital agent-​​based network-​​theory com­plex­ol­ogy nudge-​​targets
  • [1201.5426] Con­straint Prop­a­ga­tion as Infor­ma­tion Maximization

    “Dana Scott used the par­tial order among par­tial func­tions for his math­e­mat­i­cal model of recur­sively defined func­tions. He inter­preted the par­tial order as one of infor­ma­tion con­tent. In this paper we elab­o­rate on Scott’s sug­ges­tion of regard­ing com­pu­ta­tion as a process of infor­ma­tion max­i­miza­tion by apply­ing it to the solu­tion of con­straint sat­is­fac­tion prob­lems. Here the method of con­straint prop­a­ga­tion can be inter­preted as decreas­ing uncer­tainty about the solu­tion — that is, as gain in infor­ma­tion about the solu­tion. As illus­tra­tive exam­ple we choose numer­i­cal con­straint sat­is­fac­tion prob­lems to be solved by inter­val con­straints. To facil­i­tate this approach to con­straint solv­ing we for­mu­late con­straint sat­is­fac­tion prob­lems as for­mu­las in pred­i­cate logic. This neces­si­tates extend­ing the usual seman­tics for pred­i­cate logic so that mean­ing is assigned not only to sen­tences but also to for­mu­las with free variables.”

    computer-​​science quite-​​interesting constraint-​​processing computational-​​methods
  • [1201.4459] An effi­cient par­al­lel algo­rithm for the longest path prob­lem in meshes

    “In this paper, first we give a sequen­tial linear-​​time algo­rithm for the longest path prob­lem in meshes. This algo­rithm can be con­sid­ered as an improve­ment of [13]. Then based on this sequen­tial algo­rithm, we present a constant-​​time par­al­lel algo­rithm for the prob­lem which can be run on every par­al­lel machine.”

    algo­rithms graph-​​theory computational-​​complexity nudge-​​targets
  • [1201.4417] Insta­bil­i­ties and Pat­terns in Cou­pled Reaction-​​Diffusion Layers

    “We study insta­bil­i­ties and pat­tern for­ma­tion in reaction-​​diffusion lay­ers that are dif­fu­sively cou­pled. For two-​​layer sys­tems of iden­ti­cal two-​​component reac­tions, we ana­lyze the sta­bil­ity of homo­ge­neous steady states by exploit­ing the block sym­met­ric struc­ture of the lin­ear prob­lem. There are eight pos­si­ble pri­mary bifur­ca­tion sce­nar­ios, includ­ing a Turing-​​Turing bifur­ca­tion that involves two dis­parate length scales whose ratio may be tuned via the inter-​​layer cou­pling. For sys­tems of $n$-component lay­ers and non-​​identical lay­ers, the lin­ear problem’s block form allows approx­i­mate decom­po­si­tion into lower-​​dimensional lin­ear prob­lems if the cou­pling is suf­fi­ciently weak. As an exam­ple, we apply these results to a two-​​layer Brus­se­la­tor sys­tem. The com­pet­ing length scales engi­neered within the lin­ear prob­lem are read­ily appar­ent in numer­i­cal sim­u­la­tions of the full sys­tem. Select­ing a $sqrt{2}$:1 length scale ratio pro­duces an unusual steady square pattern.”

    cute emergent-​​design pattern-​​formation com­plex­ol­ogy nudge-​​targets nonlinear-​​dynamics
  • [1201.4737] Pro­duc­tion Sys­tem Rules as Pro­tein Com­plexes from Genetic Reg­u­la­tory Networks

    “This short paper intro­duces a new way by which to design pro­duc­tion sys­tem rules. An indi­rect encod­ing scheme is pre­sented which views such rules as pro­tein com­plexes pro­duced by the tem­po­ral behav­iour of an arti­fi­cial genetic reg­u­la­tory net­work. This ini­tial study begins by using a sim­ple Boolean reg­u­la­tory net­work to pro­duce tra­di­tional ternary-​​encoded rules before mov­ing to a fuzzy vari­ant to pro­duce real-​​valued rules. Com­pet­i­tive per­for­mance is shown with related genetic reg­u­la­tory net­works and rule-​​based sys­tems on bench­mark problems.”

    evolutionary-​​algorithms production-​​systems computer-​​science emergent-​​design

Items of some interest…

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

  • [1110.0585] Dis­crim­i­nately Decreas­ing Dis­crim­inabil­ity with Learned Image Filters

    “In machine learn­ing and com­puter vision, input images are often fil­tered to increase data dis­crim­inabil­ity. In some sit­u­a­tions, how­ever, one may wish to pur­posely decrease dis­crim­inabil­ity of one clas­si­fi­ca­tion task (a “dis­trac­tor” task), while simul­ta­ne­ously pre­serv­ing infor­ma­tion rel­e­vant to another (the task-​​of-​​interest): For exam­ple, it may be impor­tant to mask the iden­tity of per­sons con­tained in face images before sub­mit­ting them to a crowd­sourc­ing site (e.g., Mechan­i­cal Turk) when label­ing them for cer­tain facial attrib­utes. Another exam­ple is inter-​​dataset gen­er­al­iza­tion: when train­ing on a dataset with a par­tic­u­lar covari­ance struc­ture among mul­ti­ple attrib­utes, it may be use­ful to sup­press one attribute while pre­serv­ing another so that a trained clas­si­fier does not learn spu­ri­ous cor­re­la­tions between attrib­utes. In this paper we present an algo­rithm that finds opti­mal fil­ters to give high dis­crim­inabil­ity to one task while simul­ta­ne­ously giv­ing low dis­crim­inabil­ity to a dis­trac­tor task. We present results show­ing the effec­tive­ness of the pro­posed tech­nique on both sim­u­lated data and nat­ural face images.”

    machine-​​learning data-​​preparation fil­ter­ing algo­rithms nudge-​​targets
  • [1110.5183] Dif­fu­sion of Infor­ma­tion in Robot Swarms

    “This work is devoted to com­mu­ni­ca­tion approaches, which spread infor­ma­tion in robot swarms. These mech­a­nisms are use­ful for large-​​scale sys­tems and also for such cases when a lim­ited com­mu­ni­ca­tion equip­ment does not allow rout­ing of infor­ma­tion pack­ages. We focus on two approaches such as vir­tual fields and epi­demic algo­rithms, dis­cuss sev­eral aspects of hard­ware imple­men­ta­tion and demon­strate exper­i­ments per­formed with micro­ro­bots “Jasmine”.”

    agent-​​based swarms com­mu­ni­ca­tion complex-​​systems epi­demi­ol­ogy dynamical-​​systems exper­i­ment
  • [1109.5229] Dis­trib­uted Algo­rithms for Opti­mal Power Flow Problem

    “Opti­mal power flow (OPF) is an impor­tant prob­lem for power gen­er­a­tion and it is in gen­eral non-​​convex. With the employ­ment of renew­able energy, it will be desir­able if OPF can be solved very effi­ciently so its solu­tion can be used in real time. With some spe­cial net­work struc­ture, e.g. trees, the prob­lem has been shown to have a zero dual­ity gap and the con­vex dual prob­lem yields the opti­mal solu­tion. In this paper, we pro­pose a pri­mal and a dual algo­rithm to coor­di­nate the smaller sub­prob­lems decom­posed from the con­vex­i­fied OPF. We can arrange the sub­prob­lems to be solved sequen­tially and cumu­la­tively in a cen­tral node or solved in par­al­lel in dis­trib­uted nodes. We test the algo­rithms on IEEE radial dis­tri­b­u­tion test feed­ers, some ran­dom tree-​​structured net­works, and the IEEE trans­mis­sion sys­tem bench­marks. Sim­u­la­tion results show that the com­pu­ta­tion time can be improved dra­mat­i­cally with our algo­rithms over the cen­tral­ized approach of solv­ing the prob­lem with­out decom­po­si­tion, espe­cially in tree-​​structured prob­lems. The com­pu­ta­tion time grows lin­early with the prob­lem size with the cumu­la­tive approach while the dis­trib­uted one can have size-​​independent com­pu­ta­tion time.”

    operations-​​research algo­rithms network-​​theory infra­struc­ture com­po­si­tion nudge-​​targets
  • Coupon collector’s prob­lem — Wikipedia, the free encyclopedia

    “In prob­a­bil­ity the­ory, the coupon collector’s prob­lem describes the “col­lect all coupons and win” con­tests. It asks the fol­low­ing ques­tion: Sup­pose that there are n coupons, from which coupons are being col­lected with replace­ment. What is the prob­a­bil­ity that more than t sam­ple tri­als are needed to col­lect all n coupons? An alter­na­tive state­ment is: Given n coupons, how many coupons do you expect you need to draw with replace­ment before hav­ing drawn each coupon at least once. The math­e­mat­i­cal analy­sis of the prob­lem reveals that the expected num­ber of tri­als needed grows as Θ(nlog(n)). For exam­ple, when n = 50 it takes about 225 tri­als to col­lect all 50 coupons.”

    probability-​​theory slic-​​representation nudge-​​targets
  • [1107.2379] Data Sta­bil­ity in Clus­ter­ing: A Closer Look

    “This paper con­sid­ers the model intro­duced by Bilu and Linial (2010), who study prob­lems for which the opti­mal clus­ter­ing does not change when the dis­tances are per­turbed by mul­ti­plica­tive fac­tors. They show that even when a prob­lem is NP-​​hard, it is some­times pos­si­ble to obtain polynomial-​​time algo­rithms for instances resilient to large per­tur­ba­tions, e.g. on the order of $O(sqrt{n})$ for max-​​cut clus­ter­ing. Awasthi et al. (2010) extend this line of work by con­sid­er­ing center-​​based objec­tives, and Bal­can and Liang (2011) con­sider the $k$-median and min-​​sum objec­tives, giv­ing effi­cient algo­rithms for instances resilient to cer­tain con­stant mul­ti­plica­tive per­tur­ba­tions. Here, we are moti­vated by the ques­tion of to what extent these assump­tions can be relaxed while allow­ing for effi­cient algo­rithms. We show there is lit­tle room to improve these results by giv­ing NP-​​hardness lower bounds for both the $k$-median and min-​​sum objec­tives. On the other hand, we show that mul­ti­plica­tive resilience para­me­ters, even only on the order of $Theta(1)$, can be so strong as to make the clus­ter­ing prob­lem triv­ial, and we exploit these assump­tions to present a sim­ple one pass stream­ing algo­rithm for the $k$-median objec­tive. We also con­sider a model of addi­tive per­tur­ba­tions and give a cor­re­spon­dence between addi­tive and mul­ti­plica­tive notions of sta­bil­ity. Our results pro­vide a close exam­i­na­tion of the con­se­quences of assum­ing, even con­stant, sta­bil­ity in data.”

    clus­ter­ing sta­tis­tics algo­rithms robust­ness nudge-​​targets
  • [1109.5641] Energy-​​Aware Load Bal­anc­ing in Con­tent Deliv­ery Networks

    “Internet-​​scale dis­trib­uted sys­tems such as con­tent deliv­ery net­works (CDNs) oper­ate hun­dreds of thou­sands of servers deployed in thou­sands of data cen­ter loca­tions around the globe. Since the energy costs of oper­at­ing such a large IT infra­struc­ture are a sig­nif­i­cant frac­tion of the total oper­at­ing costs, we argue for redesign­ing CDNs to incor­po­rate energy opti­miza­tions as a first-​​order prin­ci­ple. We pro­pose tech­niques to turn off CDN servers dur­ing peri­ods of low load while seek­ing to bal­ance three key design goals: max­i­mize energy reduc­tion, min­i­mize the impact on client-​​perceived ser­vice avail­abil­ity (SLAs), and limit the fre­quency of on-​​off server tran­si­tions to reduce wear-​​and-​​tear and its impact on hard­ware reli­a­bil­ity. We pro­pose an opti­mal offline algo­rithm and an online algo­rithm to extract energy sav­ings both at the level of local load bal­anc­ing within a data cen­ter and global load bal­anc­ing across data cen­ters. We eval­u­ate our algo­rithms using real pro­duc­tion work­load traces from a large com­mer­cial CDN. Our results show that it is pos­si­ble to reduce the energy con­sump­tion of a CDN by more than 55% while ensur­ing a high level of avail­abil­ity that meets cus­tomer SLA require­ments and incur­ring an aver­age of one on-​​off tran­si­tion per server per day. Fur­ther, we show that keep­ing even 10% of the servers as hot spares helps absorb load spikes due to global flash crowds with lit­tle impact on avail­abil­ity SLAs. Finally, we show that redis­trib­ut­ing load across prox­i­mal data cen­ters can enhance ser­vice avail­abil­ity sig­nif­i­cantly, but has only a mod­est impact on energy savings.”

    algo­rithms load-​​balancing infra­struc­ture nudge-​​targets operations-​​research
  • [1110.0264] Face Recog­ni­tion using Opti­mal Rep­re­sen­ta­tion Ensemble

    “Recently, the face rec­og­niz­ers based on lin­ear rep­re­sen­ta­tions have been shown to deliver state-​​of-​​the-​​art per­for­mance. In real-​​world appli­ca­tions, how­ever, face images usu­ally suf­fer from expres­sions, dis­guises and ran­dom occlu­sions. The prob­lem­atic facial parts under­mine the valid­ity of the linear-​​subspace assump­tion and thus the recog­ni­tion per­for­mance dete­ri­o­rates sig­nif­i­cantly. In this work, we address the prob­lem in a learning-​​inference-​​mixed fash­ion. By observ­ing that the linear-​​subspace assump­tion is more reli­able on cer­tain face patches rather than on the holis­tic face, some Bayesian Patch Rep­re­sen­ta­tions (BPRs) are ran­domly gen­er­ated and inter­preted accord­ing to the Bayes’ the­ory. We then train an ensem­ble model over the patch-​​representations by min­i­miz­ing the empir­i­cal risk w.r.t the “leave-​​one-​​out mar­gins”. The obtained model is termed Opti­mal Rep­re­sen­ta­tion Ensem­ble (ORE), since it guar­an­tees the opti­mal­ity from the per­spec­tive of Empir­i­cal Risk Min­i­miza­tion. To han­dle the unknown pat­terns in test faces, a robust ver­sion of BPR is pro­posed by tak­ing the non-​​face cat­e­gory into con­sid­er­a­tion. Equipped with the Robust-​​BPRs, the infer­ence abil­ity of ORE is increased dra­mat­i­cally and sev­eral record-​​breaking accu­ra­cies (99.9% on Yale-​​B and 99.5% on AR) and desir­able effi­cien­cies (below 20 ms per face in Mat­lab) are achieved. It also over­whelms other mod­u­lar heuris­tics on the faces with ran­dom occlu­sions, extreme expres­sions and dis­guises. Fur­ther­more, to accom­mo­date immense BPRs sets, a boosting-​​like algo­rithm is also derived. The boosted model, a.k.a Boosted-​​ORE, obtains sim­i­lar per­for­mance to its pro­to­type. Besides the empir­i­cal supe­ri­or­i­ties, two desir­able fea­tures of the pro­posed meth­ods, namely, the training-​​determined model-​​selection and the data-​​weight-​​free boost­ing pro­ce­dure, are also the­o­ret­i­cally verified.”

    image-​​analysis face-​​recognition algo­rithms nudge-​​targets
  • [1110.0957] Dic­tio­nary Learn­ing for Deblur­ring and Dig­i­tal Zoom

    “This paper pro­poses a novel approach to image deblur­ring and dig­i­tal zoom­ing using sparse local mod­els of image appear­ance. These mod­els, where small image patches are rep­re­sented as lin­ear com­bi­na­tions of a few ele­ments drawn from some large set (dic­tio­nary) of can­di­dates, have proven well adapted to sev­eral image restora­tion tasks. A key to their suc­cess has been to learn dic­tio­nar­ies adapted to the recon­struc­tion of small image patches. In con­trast, recent works have pro­posed instead to learn dic­tio­nar­ies which are not only adapted to data recon­struc­tion, but also tuned for a spe­cific task. We intro­duce here such an approach to deblur­ring and dig­i­tal zoom, using pairs of blurry/​sharp (or low-​​/​high-​​resolution) images for train­ing, as well as an effec­tive sto­chas­tic gra­di­ent algo­rithm for solv­ing the cor­re­spond­ing opti­miza­tion task. Although this learn­ing prob­lem is not con­vex, once the dic­tio­nar­ies have been learned, the sharp/​high-​​resolution image can be recov­ered via con­vex opti­miza­tion at test time. Exper­i­ments with syn­thetic and real data demon­strate the effec­tive­ness of the pro­posed approach, lead­ing to state-​​of-​​the-​​art per­for­mance for non-​​blind image deblur­ring and dig­i­tal zoom.”

    image-​​processing image-​​analysis algo­rithms nudge-​​targets hyper­res­o­lu­tion
  • [1110.0477] Dis­trib­uted Evo­lu­tion­ary Graph Partitioning

    “We present a novel dis­trib­uted evo­lu­tion­ary algo­rithm, KaFF­PaE, to solve the Graph Par­ti­tion­ing Prob­lem, which makes use of KaFFPa (Karl­sruhe Fast Flow Par­ti­tioner). The use of our mul­ti­level graph par­ti­tioner KaFFPa pro­vides new effec­tive crossover and muta­tion oper­a­tors. By com­bin­ing these with a scal­able com­mu­ni­ca­tion pro­to­col we obtain a sys­tem that is able to improve the best known par­ti­tion­ing results for many inputs in a very short amount of time. For exam­ple, in Walshaw’s well known bench­mark tables we are able to improve or recom­pute 76% of entries for the tables with 1%, 3% and 5% imbalance.”

    algo­rithms graph-​​theory evolutionary-​​algorithms nudge-​​targets
  • [1104.1605] Effi­cient Top-​​K Retrieval in Online Social Tag­ging Networks

    “We con­sider in this paper top-​​k query answer­ing in social tag­ging sys­tems, also known as folk­sonomies. This prob­lem requires a sig­nif­i­cant depar­ture from exist­ing, socially agnos­tic tech­niques. In a network-​​aware con­text, one can (and should) exploit the social links, which can indi­cate how users relate to the seeker and how much weight their tag­ging actions should have in the result build-​​up. We pro­pose an algo­rithm that has the poten­tial to scale to cur­rent appli­ca­tions. While the prob­lem has already been con­sid­ered in pre­vi­ous lit­er­a­ture, this was done either under strong sim­pli­fy­ing assump­tions or under choices that can­not scale to even moderate-​​size real world appli­ca­tions. We first con­sider a key aspect of the prob­lem, which is access­ing the clos­est or most rel­e­vant users for a given seeker. We describe how this can be done on the fly (with­out any pre-​​computations) for sev­eral pos­si­ble choices — arguably the most nat­ural ones — of prox­im­ity com­pu­ta­tion in a user net­work. Based on this, our top-​​k algo­rithm is sound and com­plete, while address­ing the scal­a­bil­ity issues of the exist­ing ones. Impor­tantly, our tech­nique is instance opti­mal in the case when the search relies exclu­sively on the social weight of tag­ging actions. To fur­ther reduce response times, we then con­sider direc­tions for effi­ciency by approx­i­ma­tion. Exten­sive exper­i­ments on real world data show that our tech­niques can dras­ti­cally improve the response time, with­out sac­ri­fic­ing precision.”

    folk­son­omy tag­ging network-​​theory search-​​algorithms nudge-​​targets data-​​access

Items of some interest…

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