“… As a depar­ture from such tra­di­tional trust mod­els, we pro­pose a generic, machine learn­ing approach based trust frame­work where an agent uses its own pre­vi­ous trans­ac­tions (with other agents) to build a knowl­edge base, and uti­lize this to assess the trust­wor­thi­ness of a trans­ac­tion based on asso­ci­ated fea­tures, which are capa­ble of dis­tin­guish­ing suc­cess­ful trans­ac­tions from unsuc­cess­ful ones. These fea­tures are har­nessed using appro­pri­ate machine learn­ing algo­rithms to extract rela­tion­ships between the poten­tial trans­ac­tion and pre­vi­ous transactions.…”

“We study the 2-​​dimensional vec­tor pack­ing prob­lem, which is a gen­er­al­iza­tion of the clas­si­cal bin pack­ing prob­lem where each item has 2 dis­tinct weights and each bin has 2 cor­re­spond­ing capac­i­ties. The goal is to group items into min­i­mum num­ber of bins, with­out vio­lat­ing the bin capac­ity constraints.…”

“Thin­ning of char­ac­ter images is a big chal­lenge. Removal of strokes or defor­mi­ties in thin­ning is a dif­fi­cult problem.…”

“… We exam­ine here how the algo­rithm of phyl­lotaxis could con­tribute to the analy­sis of the struc­ture of col­la­gen fib­rils. Such an algo­rithm indeed leads to orga­ni­za­tions giv­ing to each ele­ment of the assem­bly the most homo­ge­neous and isotropic dense envi­ron­ment in a sit­u­a­tion of cylin­dri­cal sym­me­try. The scat­tered inten­sity expected from a phyl­lotac­tic dis­tri­b­u­tion of triple helices in col­la­gen fib­rils well agrees with the major fea­tures observed along the equa­to­r­ial direc­tion of their X ray pat­terns. Fol­low­ing this approach, the aggre­ga­tion of triple helices in fib­rils should be con­sid­ered within the frame of soft con­densed mat­ter stud­ies rather than that of mol­e­c­u­lar crys­tal studies.”

“Even when the num­bers of non-​​zero entries per column/​row in the mea­sure­ment matri­ces are lim­ited to $O(1)$, numer­i­cal exper­i­ments indi­cate that the algo­rithm can still typ­i­cally recover the orig­i­nal sig­nal per­fectly with an $O(N)$ com­pu­ta­tional cost per update as well if the den­sity $\rho$ of non-​​zero entries of the sig­nal is lower than a cer­tain crit­i­cal value $\rho_{\rm th}(\alpha)$ as $N,M \to \infty$.”

“We inves­ti­gate the evo­lu­tion­ary dynam­ics of an ide­alised model for the robust self-​​assembly of two-​​dimensional struc­tures called poly­omi­noes. The model includes rules that encode inter­ac­tions between sets of square tiles that drive the self-​​assembly process. The rela­tion­ship between the model’s rule set and its result­ing self-​​assembled struc­ture can be viewed as a genotype-​​phenotype map and incor­po­rated into a genetic algorithm.”

“In this work we intro­duce a new lin­ear time com­pres­sion algo­rithm, called “Re-​​pair for Trees”, which com­presses ranked ordered trees using lin­ear straight-​​line context-​​free tree gram­mars. Such gram­mars gen­er­al­ize straight-​​line context-​​free string gram­mars and allow basic tree oper­a­tions, like tra­ver­sal along edges, to be exe­cuted with­out prior decom­pres­sion. Our algo­rithm can be con­sid­ered as a gen­er­al­iza­tion of the “Re-​​pair” algo­rithm devel­oped by N. Jes­per Lars­son and Alis­tair Mof­fat in 2000. The lat­ter algo­rithm is a dictionary-​​based com­pres­sion algo­rithm for strings. We also intro­duce a suc­cinct cod­ing which is spe­cial­ized in fur­ther com­press­ing the gram­mars gen­er­ated by our algo­rithm. This is accom­plished with­out loos­ing the abil­ity do directly exe­cute queries on this com­pressed rep­re­sen­ta­tion of the input tree.…”

“Until I started using Spec­i­fi­ca­tion Work­shops as the name for a col­lab­o­ra­tive meet­ing about accep­tance tests, it was very hard to con­vince busi­ness users to par­tic­i­pate. But a sim­ple change in nam­ing made the prob­lem go away.”

“Weather data didn’t come to be because of an Open Gov­ern­ment Direc­tive. It wasn’t cre­ated because of a White House man­date. Gov­ern­ment did not release the data and then enter­pris­ing peo­ple built com­pa­nies on top of it. It’s more accu­rate to make the argu­ment that we have a national weather ser­vice because of one man’s deep desire to keep his job and to get pro­moted to colonel in the Army. It could be a vast net­work of lob­by­ists to help that man get pro­moted, or the vast net­work of lob­by­ists from ship­ping com­pa­nies try­ing to get access to data already being cre­ated. Or it could be that it was just pretty obvi­ous that access to weather data would save lives.”

“While com­pany fil­ings and reg­u­la­tions may not be the most glam­orous parts of your startup, they’re absolutely crit­i­cal to the suc­cess of your busi­ness and safety of your per­sonal sav­ings. Here’s a quick run­down of the laws and reg­u­la­tions you need to con­sider when cre­at­ing a startup. Of course, depend­ing on your type of busi­ness, hir­ing a tax accoun­tant or good attor­ney with spe­cific expe­ri­ence in your indus­try can go a long way to help­ing you steer clear of trouble.”

“We study, using sim­u­lated exper­i­ments inspired by thin film mag­netic domain pat­terns, the fea­si­bil­ity of phase retrieval in X-​​ray dif­frac­tive imag­ing in the pres­ence of intrin­sic charge scat­ter­ing given only photon-​​shot-​​noise lim­ited dif­frac­tion data. We detail a recon­struc­tion algo­rithm to recover the sample’s mag­ne­ti­za­tion dis­tri­b­u­tion under such con­di­tions, and com­pare its per­for­mance with that of Fourier trans­form holog­ra­phy. Con­cern­ing the design of future exper­i­ments, we also chart out the recon­struc­tion lim­its of dif­frac­tive imag­ing when pho­ton– shot-​​noise and the inten­sity of charge scat­ter­ing noise are inde­pen­dently var­ied. This work is directly rel­e­vant to the time-​​resolved imag­ing of mag­netic dynam­ics using coher­ent and ultra­fast radi­a­tion from X-​​ray free elec­tron lasers and also to broader classes of dif­frac­tive imag­ing exper­i­ments which suf­fer noisy data, miss­ing data or both.”

“We have shown that there exists a large ensem­ble of min­i­mal Boolean net­works that show reli­able and robust dynam­ics. The net­works are min­i­mal in the respect that the num­ber of con­nec­tions of a node is not larger than nec­es­sary for obtain­ing a desired reli­able tra­jec­tory. A reli­able tra­jec­tory is an attrac­tor of the dynam­ics of the net­work that does not change when the update sched­ule is changed or ran­dom­ized. This means that under par­al­lel update, at each time step only one node changes its state. The reli­able tra­jec­to­ries were cho­sen at ran­dom, given a fixed aver­age num­ber of flips per node. High robust­ness was achieved by using an evo­lu­tion­ary algo­rithm that mod­i­fies the update func­tions and that accepts only those changes that do not decrease robustness.…”

“Our 2.546-approximation is quite sim­ple. The per­for­mance guar­an­tee is based on a sim­ple area argu­ment. This gives rise to the fol­low­ing ques­tion: what is the small­est square that suf­fices for pack­ing any set of cir­cles of total area 1? We believe the worst-​​case may very well be shown in Fig­ure 13, which yields a lower bound of 1.471299… We believe there are rel­a­tively easy ways to improve the upper bound.”

“In this paper we present a tech­nique for fusion of opti­cal and ther­mal face images based on image pixel fusion approach. Out of sev­eral fac­tors, which affect face recog­ni­tion per­for­mance in case of visual images, illu­mi­na­tion changes are a sig­nif­i­cant fac­tor that needs to be addressed. Ther­mal images are bet­ter in han­dling illu­mi­na­tion con­di­tions but not very con­sis­tent in cap­tur­ing tex­ture details of the faces. Other fac­tors like sun­glasses, beard, mous­tache etc also play active role in adding com­pli­ca­cies to the recog­ni­tion process. Fusion of ther­mal and visual images is a solu­tion to over­come the draw­backs present in the indi­vid­ual ther­mal and visual face images.…”

“OpenCV (Open Source Com­puter Vision) is a library of pro­gram­ming func­tions for real time com­puter vision.

OpenCV is released under a BSD license, it is free for both aca­d­e­mic and com­mer­cial use.
The library has >500 opti­mized algo­rithms (see fig­ure below). It is used around the world, has >2M down­loads and >40K peo­ple in the user group. Uses range from inter­ac­tive art, to mine inspec­tion, stitch­ing maps on the web on through advanced robotics.”

“We study the dynam­ics of the Nam­ing Game [Baronchelli et al., (2006) J. Stat. Mech.: The­ory Exp. P06014] in empir­i­cal social net­works. This styl­ized agent-​​based model cap­tures essen­tial fea­tures of agree­ment dynam­ics in a net­work of autonomous agents, cor­re­spond­ing to the devel­op­ment of shared clas­si­fi­ca­tion schemes in a net­work of arti­fi­cial agents or opin­ion spread­ing and social dynam­ics in social net­works. Our study focuses on the impact that com­mu­ni­ties in the under­ly­ing social graphs have on the out­come of the agree­ment process. We find that net­works with strong com­mu­nity struc­ture hin­der the sys­tem from reach­ing global agree­ment; the evo­lu­tion of the Nam­ing Game in these net­works main­tains clus­ters of coex­ist­ing opin­ions indef­i­nitely. Fur­ther, we inves­ti­gate agent-​​based net­work strate­gies to facil­i­tate con­ver­gence to global consensus.”

“Neu­ronal activ­ity arises from an inter­ac­tion between ongo­ing fir­ing gen­er­ated spon­ta­neously by neural cir­cuits and responses dri­ven by exter­nal stim­uli. Using mean-​​field analy­sis, we ask how a neural net­work that intrin­si­cally gen­er­ates chaotic pat­terns of activ­ity can remain sen­si­tive to extrin­sic input. We find that inputs not only drive net­work responses, they also actively sup­press ongo­ing activ­ity, ulti­mately lead­ing to a phase tran­si­tion in which chaos is com­pletely elim­i­nated. The crit­i­cal input inten­sity at the phase tran­si­tion is a non-​​monotonic func­tion of stim­u­lus fre­quency, reveal­ing a “res­o­nant” fre­quency at which the input is most effec­tive at sup­press­ing chaos even though the power spec­trum of the spon­ta­neous activ­ity peaks at zero and falls expo­nen­tially. A pre­dic­tion of our analy­sis is that the vari­ance of neural responses should be most strongly sup­pressed at fre­quen­cies match­ing the range over which many sen­sory sys­tems operate.”

“Devel­op­ing large-​​scale dis­trib­uted appli­ca­tions can be a daunt­ing task. object-​​based envi­ron­ments have attempted to alle­vi­ate prob­lems by pro­vid­ing dis­trib­uted objects that look like local objects. We advo­cate that this approach has actu­ally only made mat­ters worse, as the devel­oper needs to be aware of many intri­cate inter­nal details in order to ade­quately han­dle par­tial fail­ures. The result is an increase of appli­ca­tion com­plex­ity. We present an alter­na­tive in which dis­tri­b­u­tion trans­parency is less­ened in favor of clearer seman­tics. In par­tic­u­lar, we argue that a devel­oper should always be offered the unam­bigu­ous seman­tics of local objects, and that dis­tri­b­u­tion comes from copy­ing those objects to where they are needed. We claim that it is often suf­fi­cient to pro­vide only small, immutable objects, along with facil­i­ties to group objects into clusters.”

“The task of image restra­tion is to find the spa­tial cor­re­spon­dence of two or more given images. In this paper we assume that the cor­re­spon­dence is given either by an Euclid­ean, or by an affine volume-​​preserving trans­for­ma­tion. Since the reg­is­tra­tion prob­lem can be seen as an opti­miza­tion prob­lem on a finite dimen­sional Lie group, we use a recently devel­oped frame­work of approximate-​​Newton meth­ods on man­i­folds, which leads to locally qua­drat­i­cally con­ver­gent algo­rithms. To reduce numer­i­cal costs, we present two strate­gies: One makes use of the quasi Monte Carlo Method and the other ends up with an algo­rithm act­ing on spline func­tion spaces. An exten­sion for multi-​​modal image reg­is­tra­tion is given as well.”

“l1-​​minimization refers to find­ing the min­i­mum l1-​​norm solu­tion to an under­de­ter­mined lin­ear sys­tem b=Ax. It has recently received much atten­tion, mainly moti­vated by the new com­pres­sive sens­ing the­ory that shows that under quite gen­eral con­di­tions the min­i­mum l1-​​norm solu­tion is also the spars­est solu­tion to the sys­tem of lin­ear equa­tions. Although the under­ly­ing prob­lem is a lin­ear pro­gram, con­ven­tional algo­rithms such as interior-​​point meth­ods suf­fer from poor scal­a­bil­ity for large-​​scale real world prob­lems. A num­ber of accel­er­ated algo­rithms have been recently pro­posed that take advan­tage of the spe­cial struc­ture of the l1-​​minimization prob­lem. In this paper, we pro­vide a com­pre­hen­sive review of five rep­re­sen­ta­tive approaches, namely, Gra­di­ent Pro­jec­tion, Homo­topy, Iter­a­tive Shrinkage-​​Thresholding, Prox­i­mal Gra­di­ent, and Aug­mented Lagrange Multiplier. …”

“Sparse data mod­els, where data is assumed to be well rep­re­sented as a lin­ear com­bi­na­tion of a few ele­ments from a dic­tio­nary, have gained con­sid­er­able atten­tion in recent years, and their use has led to state-​​of-​​the-​​art results in many sig­nal and image pro­cess­ing tasks. It is now well under­stood that the choice of the spar­sity reg­u­lar­iza­tion term is crit­i­cal in the suc­cess of such mod­els. Based on a code­length min­i­miza­tion inter­pre­ta­tion of sparse cod­ing, and using tools from uni­ver­sal cod­ing the­ory, we pro­pose a frame­work for design­ing spar­sity reg­u­lar­iza­tion terms which have the­o­ret­i­cal and prac­ti­cal advan­tages when com­pared to the more stan­dard l0 or l1 ones. The pre­sen­ta­tion of the frame­work and the­o­ret­i­cal foun­da­tions is com­ple­mented with exam­ples that show its prac­ti­cal advan­tages in image denois­ing, zoom­ing and classification.”

“In the present arti­cle we empha­size the impor­tance of mod­el­ing time in the con­text of agent-​​based mod­els. To this end, we present a (selec­tive) sur­vey of the Cel­lu­lar Automata-​​literature on updat­ing and draw par­al­lels to the issue of agent acti­va­tion in agent-​​based mod­els. By means of two sim­ple mod­els, Schelling’s seg­re­ga­tion model and Epstein’s demo­graphic prisoner’s dilemma we inves­ti­gate the influ­ence of choos­ing dif­fer­ent regimes of agent acti­va­tion. Our exper­i­ments indi­cate that tim­ing is not a crit­i­cal issue for very sim­ple mod­els but bears huge influ­ence on model behav­ior and results as soon as the degree of com­plex­ity increases only so slightly. After a brief review of the way com­monly used ABM sim­u­la­tion envi­ron­ments han­dle the issue of tim­ing, we draw some ten­ta­tive con­clu­sions about the impor­tance of tim­ing and the need for more research towards that direc­tion, sim­i­lar to the con­certed effort on updat­ing in cel­lu­lar automata.”

“Tran­sient alge­bra is a multi-​​valued alge­bra for haz­ard detec­tion in gate cir­cuits. Sequences of alter­nat­ing 0’s and 1’s, called tran­sients, rep­re­sent sig­nal val­ues, and gates are mod­eled by exten­sions of boolean func­tions to tran­sients. For­mu­las for com­put­ing the out­put tran­sient of a gate from the input tran­sients are known for NOT, AND, OR} and XOR gates and their com­ple­ments, but, in gen­eral, even the prob­lem of decid­ing whether the length of the out­put tran­sient exceeds a given bound is NP-​​complete. We pro­pose a method of eval­u­at­ing exten­sions of gen­eral boolean func­tions. We intro­duce and study a class of func­tions with the fol­low­ing prop­erty: Instead of eval­u­at­ing an exten­sion of a boolean func­tion on a given set of tran­sients, it is pos­si­ble to get the same value by using tran­sients derived from the given ones, but hav­ing length at most 3. We prove that all func­tions of three vari­ables, as well as cer­tain other func­tions, have this prop­erty, and can be effi­ciently evaluated.”

“We define a two-​​step learner for RFSAs based on an obser­va­tion table by using an algo­rithm for min­i­mal DFAs to build a table for the rever­sal of the lan­guage in ques­tion and show­ing that we can derive the min­i­mal RFSA from it after some sim­ple mod­i­fi­ca­tions. We com­pare the algo­rithm to two other table-​​based ones of which one (by Bol­lig et al. 2009) infers a RFSA directly, and the other is another two-​​step learner pro­posed by the author. We focus on the cri­te­rion of query complexity.”

“…The model is based on the emer­gent prop­er­ties of generic genetic net­works, it does not refer to spe­cific con­trol cir­cuits and it can there­fore hold for a wide class of lin­eages. The model points to a pecu­liar role of cel­lu­lar noise in dif­fer­en­ti­a­tion, which has never been hypoth­e­sized so far, and leads to non triv­ial pre­dic­tions which could be sub­ject to exper­i­men­tal testing.”

“This paper presents a novel type-​​2 Fuzzy logic Sys­tem to define the Shape of a facial com­po­nent with the crisp out­put. This work is the part of our main research effort to design a sys­tem (called FASY) which offers a novel face con­struc­tion approach based on the tex­tual descrip­tion and also extracts and ana­lyzes the facial com­po­nents from a face image by an effi­cient tech­nique. The Fuzzy model, designed in this paper, takes crisp value of width and height of a facial com­po­nent and pro­duces the crisp value of Shape for dif­fer­ent facial com­po­nents. This method is designed using Mat­lab 6.5 and Visual Basic 6.0 and tested with the facial com­po­nents extracted from 200 male and female face images of dif­fer­ent ages from dif­fer­ent face databases.”

“We con­sider a gen­er­al­iza­tion of the Gabriel graph, the wit­ness Gabriel graph. Given a set of ver­tices P and a set of wit­nesses W in the plane, there is an edge ab between two points of P in the wit­ness Gabriel graph GG-(P,W) if and only if the closed disk with diam­e­ter ab does not con­tain any wit­ness point (besides pos­si­bly a and/​or b). We study sev­eral prop­er­ties of the wit­ness Gabriel graph, both as a prox­im­ity graph and as a new tool in graph drawing.”

“We use com­puter sim­u­la­tion to study crystal-​​forming model pro­teins equipped with inter­ac­tions that are both ori­en­ta­tion­ally spe­cific and non­spe­cific. Dis­tinct dynam­i­cal path­ways of crys­tal for­ma­tion can be selected by tun­ing the strengths of these inter­ac­tions. When the non­spe­cific inter­ac­tion is strong, liq­uid­like clus­ter­ing can pre­cede crys­tal­liza­tion; when it is weak, growth can pro­ceed via ordered nuclei. Crys­tal yields are in cer­tain para­me­ter regimes enhanced by the non­spe­cific inter­ac­tion, even though it pro­motes asso­ci­a­tion with­out local crys­talline order. Our results sug­gest that equip­ping nanoscale com­po­nents with weak non­spe­cific inter­ac­tions (such as deple­tion attrac­tions) can alter both their dynam­i­cal path­way of assem­bly and opti­mize the yield of the result­ing material.”

“Many com­plex sys­tems present an intrin­sic bipar­tite nature and are often described and mod­eled in terms of net­works [1–5]. Exam­ples include movies and actors [1, 2, 4], authors and sci­en­tific papers [6–9], email accounts and emails [10], plants and ani­mals that pol­li­nate them [11, 12]. Bipar­tite net­works are often very het­ero­ge­neous in the num­ber of rela­tion­ships that the ele­ments of one set estab­lish with the ele­ments of the other set. … Here we intro­duce an unsu­per­vised method to sta­tis­ti­cally val­i­date each link of the pro­jected net­work against a null hypoth­e­sis tak­ing into account the het­ero­gene­ity of the sys­tem. We apply our method to three dif­fer­ent sys­tems…. In all these sys­tems, both dif­fer­ent in size and level of het­ero­gene­ity, we find that our method is able to detect net­work struc­tures which are infor­ma­tive about the system…”

“Work out which bits of the sys­tem you know least about. Cre­ate the sce­nar­ios and have con­ver­sa­tions around those bits of the sys­tem. You don’t have to grow the sys­tem from the begin­ning – you can pick any point you like! Which bits of the sys­tem make you most uncom­fort­able? Which bits make your stake­hold­ers most uncomfortable?”

“So where does this gulf of expe­ri­ences come from, why is cucum­ber loved by some and hated by oth­ers. At the risk of over-​​generalisation and mis­char­ac­ter­i­sa­tion I recently came up with a the­ory: the cucum­ber detrac­tors are not using cuke the way it was intended.”

“We tried a vari­ant of this pro­gram start­ing in 2002 with a more solid econ­omy and we are still try­ing to recover from how that movie ended. Ein­stein defined insan­ity as doing the same thing over and over again and expect­ing dif­fer­ent results. And since the finan­cial sec­tor prof­ited so hand­somely from this exer­cise the last time around, they have every rea­son to encour­age this insanity.”

“In this work we inves­ti­gate a novel approach to han­dle the chal­lenges of face recog­ni­tion, which includes rota­tion, scale, occlu­sion, illu­mi­na­tion etc. Here, we have used ther­mal face images as those are capa­ble to min­i­mize the affect of illu­mi­na­tion changes and occlu­sion due to mous­tache, beards, adorn­ments etc. The pro­posed approach reg­is­ters the train­ing and test­ing ther­mal face images in polar coor­di­nate, which is capa­ble to han­dle com­pli­ca­cies intro­duced by scal­ing and rota­tion. Line fea­tures are extracted from ther­mal polar images and fea­ture vec­tors are con­structed using these line. Fea­ture vec­tors thus obtained passes through prin­ci­pal com­po­nent analy­sis (PCA) for the dimen­sion­al­ity reduc­tion of fea­ture vectors.…”

“In 1961 Her­bert Simon and Albert Ando pub­lished the the­ory behind the long-​​term behav­ior of a dynam­i­cal sys­tem that can be described by a nearly com­pletely decom­pos­able matrix. Over the past fifty years this the­ory has been used in a vari­ety of con­texts, includ­ing queue­ing the­ory, com­puter per­for­mance, and ecol­ogy. In all these appli­ca­tions, the struc­ture of the sys­tem is known and the point of inter­est is the var­i­ous states the sys­tem passes through on its way to some long-​​term equi­lib­rium. This paper looks at this prob­lem from the other direc­tion. That is, we develop a tech­nique for using the evo­lu­tion of the sys­tem to tell us about its ini­tial struc­ture, and we use this tech­nique to develop a new algo­rithm for data clustering.”

“Smoothie Charts is a really small chartling library designed for live stream­ing data. I built it to reduce the headaches I was get­ting from watch­ing charts jerk­ily updat­ing every sec­ond. What you’re look­ing up now is pretty much all it does. If you like that, then read on.”

“We dis­cuss the com­plex dynam­ics of a non-​​linear ran­dom net­works model, as a func­tion of the con­nec­tiv­ity k between the ele­ments of the net­work. We show that this class of net­works exhibit an order-​​chaos phase tran­si­tion for a crit­i­cal con­nec­tiv­ity k = 2. Also, we show that both, pair­wise cor­re­la­tion and com­plex­ity mea­sures are max­i­mized in dynam­i­cally crit­i­cal net­works. These results are in good agree­ment with the pre­vi­ously reported stud­ies on ran­dom Boolean net­works and ran­dom thresh­old net­works, and show once again that crit­i­cal net­works pro­vide an opti­mal coor­di­na­tion of diverse behavior.”

“The new algo­rithm by Yadava and his col­leagues goes one step fur­ther. It uses a more real­is­tic physics model of the x-​​ray source, the detec­tors, and the x-​​ray beam. Each of these three is assumed to have spe­cific diam­e­ters instead of being con­sid­ered a point or a line, Yadava says. Depend­ing on the type of scan, the tech­nique is bet­ter than ASIR at cut­ting image noise, and thus the x-​​rays can be even less intense. The researchers got high-​​quality abdomen scans of a human model using an eighth of the radi­a­tion dose of a con­ven­tional scan.”

“Because shar­ing data resulted in a cita­tion, I won­der how long will it take for Open Data advo­cates to start using this “open data cita­tion advan­tage” as an argu­ment for shar­ing data?”

“As we con­tinue to mon­i­tor the mar­kets for evi­dence of Quote Stuff­ing and Strange Sequences (Crop Cir­cles), we find that there are dozens if not hun­dreds of exam­ples to choose from on any given day. As such, this page will be updated often with charts demon­strat­ing this activity.

The com­mon theme with the charts shown on this page is they are obvi­ously all gen­er­ated in code and are algo­rith­mic. Some demon­strate bizarre price or size cycling, some demon­strate large burst of quotes in extremely short time frames and some will demon­strate both. In most cases these sequences are from a sin­gle exchange with no other exchange quot­ing in the same time frame.”

“While ana­lyz­ing HFT (High Fre­quency Trad­ing) quote counts, we were shocked to find cases where one exchange was send­ing an extremely high num­ber of quotes for one stock in a sin­gle sec­ond: as high as 5,000 quotes in 1 sec­ond! Dur­ing May 6, there were hun­dreds of times that a sin­gle stock had over 1,000 quotes from one exchange in a sin­gle sec­ond. Even more dis­turb­ing, there doesn’t seem to be any eco­nomic jus­ti­fi­ca­tion for this. In many of the cases, the bid/​offer is well out­side the National Best Bid/​Offer (NBBO). We decided to ana­lyze a hand­ful of these cases in detail and graphed the sequen­tial bid/​offers to bet­ter under­stand them. What we dis­cov­ered was a manip­u­la­tive device with desta­bi­liz­ing effect.”

“…We tested our algo­rithm on a small bench­mark graph and on a net­work of about 500 papers in infor­ma­tion sci­ence (weighted with the Salton index of bib­li­o­graphic cou­pling). In our tests, this approach results in char­ac­ter­is­tic ranges of res­o­lu­tion where a large res­o­lu­tion change does not lead to a growth of the nat­ural com­mu­nity. Such sta­ble mod­ules were also obtained by apply­ing the LFK algo­rithm but since we deter­mine com­mu­ni­ties for all res­o­lu­tion val­ues in one run, our approach is faster than the LFK ref­er­ence. And our algo­rithm reveals the hier­ar­chi­cal struc­ture of the graph more easily.”

can I have some of these par­ti­cles, please?

“The pos­si­bil­ity of micro­scopic swim­ming by extrac­tion of energy from an exter­nal flow is dis­cussed, focus­ing on the migra­tion of a sim­ple trimer across a lin­ear shear flow. The geo­met­ric prop­er­ties of swim­ming, together with the pos­si­ble gen­er­al­iza­tion to the case of a vesi­cle, are analyzed.The mech­a­nism of energy extrac­tion from the flow appears to be the gen­er­al­iza­tion to a dis­crete swim­mer of the tank-​​treading regime of a vesi­cle. The swim­mer takes advan­tage of the exter­nal flow by both extract­ing energy for swim­ming and “sail­ing” through it. The migra­tion veloc­ity is found to scale lin­early in the stroke ampli­tude, and not qua­drat­i­cally as in a qui­es­cent fluid. This effect turns out to be con­nected with the non-​​applicability of the scal­lop the­o­rem in the pres­ence of exter­nal flow fields.”

“A par­ti­cle that moves along a smooth track in a ver­ti­cal plane is influ­enced by two forces: grav­ity and nor­mal force. The force expe­ri­enced by roller coaster rid­ers is the nor­mal force, so a nat­ural ques­tion to ask is: what shape of the track gives a nor­mal force of con­stant mag­ni­tude? Here we solve this prob­lem. It turns out that the solu­tion is related to the Kepler prob­lem; the tra­jec­to­ries in veloc­ity space are conic sections.”

“Recently pop­u­lar­ized ran­dom­ized meth­ods for prin­ci­pal com­po­nent analy­sis (PCA) effi­ciently and reli­ably pro­duce nearly opti­mal accu­racy — even on par­al­lel proces­sors — unlike the clas­si­cal (deter­min­is­tic) alter­na­tives. We adapt one of these ran­dom­ized meth­ods for use with data sets that are too large to be stored in random-​​access mem­ory (RAM). (The tra­di­tional ter­mi­nol­ogy is that our pro­ce­dure works effi­ciently “out-​​of-​​core.”) We illus­trate the per­for­mance of the algo­rithm via sev­eral numer­i­cal exam­ples. For exam­ple, we report on the PCA of a data set stored on disk that is so large that less than a hun­dredth of it can fit in our computer’s RAM.”

“A story isn’t done until it is being used by real users in pro­duc­tion and has been val­i­dated to be a use­ful part of a product.”

“Although Laposky, a drafts­man by pro­fes­sion, had received a proper recog­ni­tion as a pio­neer of elec­tronic art, at no time his name has emerged in the con­text of dynam­i­cal chaos the­ory. The cir­cuits he had imple­mented for gen­er­a­tion of “oscil­lons” on the screen of a cath­ode ray tube oscil­lo­scope, remain a mys­tery. It is known that some of his thirty-​​seven cir­cuits [2] had “as many as 70 dif­fer­ent set­ting of controls”[3] and that ac-​​voltage has been used for the cir­cuit feed­ing. Our analy­sis is based on the van­ity press book­let with the still pho­tos of the fifty-​​six oscil­lons, which were exhib­ited at the San­ford Museum (Chero­kee, Iowa) in 1953 [2].”

“It wasn’t all that long ago that the media and bank­ing indus­try com­men­ta­tors would worry about the com­ing train wreck in com­mer­cial real estate. But pecu­liarly, that topic has more or less receded from view. It appears the pub­lic has only so much inter­est in bank­ing sto­ries, and the fren­zied cov­er­age of finan­cial ser­vices non-​​reform plus euro­zone sov­er­eign debt woes, which are really euro­zone bank woes, took cen­ter stage.”

“It is my opin­ion, that any­body should be able to use Couch in Rails or Sina­tra or plain Ruby appli­ca­tion as eas­ily as using ActiveRe­cord, or, maybe more impor­tantly, the highly faved Mon­goDB. Please share your opin­ion in the comments.”

“Is it any won­der that the value of a col­lege edu­ca­tion is now being ques­tioned more than it used to be? Per­haps a basic edu­ca­tion in per­sonal finance would help more peo­ple make informed deci­sions about col­lege and how to han­dle the financ­ing of that endeavor.”

“… In par­tic­u­lar, the fetishiza­tion of auton­omy, self-​​reliance, and progress through mechan­i­cal engi­neer­ing — echo­ing the desire to escape the suf­fo­cat­ing social con­di­tions back east by sim­ply run­ning away — utterly under­mine the pro­gram itself and are incom­pat­i­ble with life in a space colony (which is likely to be at a min­i­mum some­what more con­strained than life in one of the more bureau­crat­i­cally obsessive-​​compulsive Euro­pean social democ­ra­cies, and at worst will tend towards the state of North Korea in Space).”

“Look no one wants to see vio­lence in the streets, but his­tory shows that it is not only the cap­i­tal­ists that have 2^nd amend­ment reme­dies. Joe Hill may have more life in him than they like.”