Sparse Structured Associative Memories as Efficient Set-Membership Data Structures

Size: px
Start display at page:

Download "Sparse Structured Associative Memories as Efficient Set-Membership Data Structures"

Transcription

1 Sparse Strutured Assoiative Memories as Effiient Set-Membership Data Strutures Vinent Gripon Eetronis Department Tééom Bretagne, Brest, Frane Vitay Skahek Institute of Computer Siene University of Tartu, Estonia Mihae Rabbat Eetria and Computer Engineering MGi University, Montréa, Canada Abstrat We study the use of sparse strutured assoiative memories as a memory-effiient and omputationay-effiient data struture for representing a set of eements when one wishes to perform set-membership ueries and some errors fase positives are toerabe. Assoiative memories, when viewed as representing a set, enjoy a number of interesting properties, inuding that set membership ueries an be arried out even when the input uery eement is ony partiay known or is partiay orrupted. The assoiative memories onsidered here initiay proposed in [Gripon and Berrou, 0] enode the set in the edge struture of a graph. In this paper we generaize this onstrution to enode the set in the edge struture of a hypergraph. We derive bounds on the fase positive rates the probabiity that the assoiative memory erroneousy deares that an eement is in the stored set when it, in fat, was not. Interestingy, the proposed strutures enjoy many of the same properties as Boom fiters e.g., they have zero fase negative rate, the time to perform an insert and ookup does not depend on the number of eements stored, and the fase positive rate an be redued by using additiona memory for storage, whie aso offering the properties of assoiative memories aowing for ueries on partia or orrupted inputs. I. INTRODUCTION A wide variety of appiations reuire the abiity to uiky determine whether a given eement is a member of a pre-speified subset of possibe eements. For exampe, in network seurity and intrusion detetion, a paket must be uiky tested to see if it mathes patterns of maware or known attaks []. Simiary, ueries in database systems an be resoved or approximated set representations and operations e.g., ounting [], []. In genera, we onsider the probem of representing a subset S of a finite universe of eements U with U = n and S = m < n. The aim is to represent S using a data struture whih an uiky test whether a partiuar eement u U is a member of S i.e., in Θog n time, whie not reuiring substantia memory. Observe that storing S as a sorted array woud reuire Θm og n bits of memory and eah uery woud have a ompexity of Θog m og n time. Boom fiters [4], and their variations [5], are a popuar randomized data struture for representing sets in order to effiienty perform set membership ueries when some errors are toerabe. They store the subset S in suh a way that set membership ueries have no fase negatives i.e., if u S, then the fiter aways reognizes u as being in S, but fase positives may our i.e., an eement u U, u / S may be fasey deared to be a member of S. Speifiay, Boom fiters represent S using an array of b bits and k hash funtions h,..., h k, with eah h j mapping eements of U to [b] def = {,..., b}. The empty set S = is represented by setting a bits to 0. To add an eement x S, the bits indexed by {h j x} k j= are set to. One a bit is set to, it remains set at regardess of whether other eements of S aso hash to the same bit. After a eements x S have been inserted, to test whether an eement x is in S, one heks the status of the bits indexed by {h j x } k j= ; if a bits are set to one, then we deare x S, and otherwise we deare x / S. Ceary, as more eements are inserted in the data struture, more bits wi be set to, eading to more fase positives. The fase positive rate an be ontroed by appropriate hoie of the number of bits b and the number of hash funtions k. In partiuar, the fase positive rate is guaranteed to be no more than ɛ > 0 if b m og e og /ɛ bits and k = b/m og e hash funtions are used, regardess of the size of the universe. Insertions and ueries both have ompexity of Θk og n time if the hash funtions are evauated one after the other, or Θog n if the hash funtions an be evauated in parae. In this paper we onsider an aternative to Boom fiters for representing sets. Our approah buids on an assoiative memory onstrution proposed by Gripon and Berrou [6] whih an be viewed as sparse, strutured binary graphia modes with binary-vaued edge-potentias. We refer to these as sparse, strutured assoiative memories SSAMs. SSAMs enjoy many of the same properties as Boom fiters. Insertions and ueries an be performed in Θog n time. They never produe fase negatives, and their fase positive rate an be redued through the use of additiona bits of memory. In addition, SSAMs enjoy the error-orreting properties of assoiative memories; that is, they an be used to test if an eement x is in the stored set, even if a portion of the binary representation of x has been erased or orrupted. On the other hand, the anaysis of SSAMs assumes that the set S being represented is drawn uniformy from a subsets of m eements from U. The ontributions of this artie are as foows. First, we present detaied error bounds on the fase positive rate for

2 uster uster uster uster 4 Fig.. Iustration of the storing proess of the message in a graph using the prinipes desribed in [6]. SSAMs. The SSAM mode desribed in [6] represents the set S in terms of the edges of a graph see Setion II for detais. In this paper we present an extension of the SSAM mode to hypergraphs and demonstrate how this eads to improved error rates, at the ost of higher storage. We show that using a -reguar hypergraph is more effiient in terms of the amount of memory reuired when the number of messages m to be stored is in the regime Θn og n /. II. GRAPH-BASED SPARSE STRUCTURED ASSOCIATIVE MEMORIES This setion desribes the onstrution of SSAMs foowing the desription in [6]. Rea that eah eement x U an be represented using og n bits. Euivaenty, eah eement an be represented as a string of symbos from an aphabet A = [] as ong as og og n. For an eement x U, et w i x A, i [], denote the ith symbo in its representation as a string of symbos in A. An SSAM is represented as a simpe undireted graph G = V, E, and in partiuar, the set S = {x,..., x m } U of eements to be stored is enoded in the edge set of the graph. Let and be positive integers suh that og og n. The graph G has V = verties; more preisey, G is -partite with eah partition ontaining verties. Let v i,j, i [], j [], denote the jth vertex in the ith partition. Initiay the graph is empty; i.e., E =. Then, for eah eement x S, we add edges between a distint pairs of verties v i,wix, v i,w i x, where i, i [], i i. One an edge has been added to the graph for one eement x S, it remains in the graph i.e., nothing hanges if another eement x S aso inudes this edge. Observe that, after a eements have been added, eah eement x S orresponds to a iue in G. Let Cx = { v i,wix, v i,w i x V V : i, i, i i } denote the edges omprising the iue orresponding to x U. Then, after adding a eements in S to the SSAM we have that E = m i= Cx i. See Figure for an iustration. Now, given the graph G representing the set S, the SSAM deares that a given eement x is in S if a edges v i,wix, v i,w i x, i, i, i i, are in E; i.e., if Cx E. Ceary, there are no fase negatives, sine x S impies that Cx E. However, there may be fase positives. In partiuar, as S grows, more edges are added to E and the fase positive rate wi inrease. We provide bounds on the fase positive rate in the foowing setion. Before proeeding, we note that representing G reuires bits sine G is -partite and eah partition ontains verties. The operations of adding an eement to the SSAM and uerying an eement for set membership both have ompexity of Θ time sine they invove setting or heking the status of bits. III. FALSE POSITIVE RATE Next we derive bounds on the fase positive rate of the SSAMs desribed in the previous setion. We assume that the set S = {x,..., x m } is omprised of m random eements x i U euivaenty, m words in [] drawn uniformy and independenty. Let P denote the orresponding probabiity measure. To avoid triviaities, we assume that m >. Rea that there are possibe edges that may be in the graph. Eah time a word is inserted, one edge is added if it isn t aready present between eah partition of G. Therefore, after inserting m eements, the probabiity that a partiuar edge has not been added is m. The expeted density the proportion of edges present in G after a m eements are inserted is thus ρ = m. Our main resut haraterizes the error rate when, for a fixed density, we et the size of the network grow as the number of eements inserted inreases. Theorem : Let ρ and K be fixed onstants satisfying 0 < ρ < and 0 < K <, and et and be positive integers satisfying og og n and hosen suh that as. Then ρ K / 4 P Cx E x / S exp{ Θ}. 5 Remark : Note that the ondition 4 states that binary messages are spit into submessages whose number,, and ength, og, are of the same order of magnitude. The onstant K is a saing fator, the roe of whih wi beome ear in the seue. It aso foows from that the number of messages m to be stored is of the order of ρ K. Theorem states that the fase positive rate deays exponentiay in, the number of partitions in G, as ong as and are hosen to satisfy 4, whih ditates the rate at whih the size of the network shoud grow as a funtion of the number of eements whih have been added. To prove Theorem we need to bound the probabiity that G ontains a iue Cx for some eement x / S. For two eements, x, x U, et dx, x denote the Hamming distane between wx and wx, where wx = w x, w x,..., w x is the representation of the eement using symbos from

3 the aphabet A = []. Hene dx, x is a non-negative integer between 0 and. With sight abuse of notation, for a set S U, et dx, S = min x S dx, x. The proof of Theorem hinges on the foowing emma. Lemma : Let ρ and K be fixed onstants satisfying 0 < ρ < and 0 < K <, and et and be positive integers satisfying og og n and hosen suh that ρ K / 6 as. Let δ [] be an integer. Then, for x 0 suh that dx 0, S = δ, P Cx 0 E x 0 / S exp{ Θ}. 7 Proof: [Sketh of proof] For the eements {x,..., x m } omprising the set S, et C i = Cx i denote the orresponding iue. Let us aso write C 0 = Cx 0 for the eement x 0 / S. Without oss of generaity, suppose that dx 0, x = δ, and suppose that wx 0 and wx differ in the ast δ symbos, where wx = w x,..., w x. To simpify notation, et us further assume without oss of generaity that C is the set of a edges between pairs of verties in the set {v,, v,,..., v, }, and C 0 is the set of a edges between pairs of verties in the set {v,, v,,..., v δ,, v δ+,,..., v, }. In other words, C 0 differs from C in edges invoving the ast δ verties. Let E = C 0 \ C. Sine the expeted probabiity that a speifi edge is present in G is ρ, and sine v,, v δ+, / C, we have P v,, v δ+, E C ρ. Next, onsider the event that the edge v,, v δ+, E, and et C be a iue that ontains it. Then the edge v,, v δ+, is in E if it either beongs to C or if it beongs to a iue different from C and C. For the first ase, we have P v,, v δ+, C C, v,, v δ+, C =. For the seond ase, simiar to above, we obtain P v,, v δ+, E C, v,, v δ+, C, v,, v δ+, / C ρ. Continuing aong this ine of reasoning, we an bound the probabiity that eah of the edges in E are in E. Let and et V both = {v,, v,...., v δ, }, V diff = {v δ+,, v δ+,,..., v, }. Observe that E is exaty E = V both V diff V diff V diff \{v i,j, v i,j : v i,j V diff }. Therefore, there are E = δδ + δδ = δ δ + edges in E, and we obtain that P E E δ+ δ C ρ + }{{} ρ δ δ+. Rea that, by assumption, ρ is a fixed onstant and = exp{θ}. Thus, for suffiienty arge, we have P C 0 E C ρ δ+ δ δ+ δ ρ, where ρ approahes ρ as sine, from 6, grows exponentiay as. Now, fix x 0 U euivaenty, fix the iue C 0 or the word w 0 = wx []. For a random eement x U, P dx, x = δ > 0 δ δ δ. Sine <, it foows that the joint probabiity that C E, C 0 E, and dc 0, C = δ is at most δ δ+ δ ρ. δ There are m eements x,..., x m in S, and we assume eah is drawn independenty and uniformy from U. Therefore, appying the union bound, the probabiity that at east one x i is at Hamming distane δ from x 0 is at most Sine m δ δ δ+ δ ρ. ρ K /, and sine m og ρ f. Euation, we onude that the probabiity that there is an eement x i in S at distane δ from x 0 and that C 0 E is bounded from above by δ δ+ δ m ρ δ < og ρ ρ K δ +δ δ+δ. δ There are possibe ways to hoose x 0. By the union bound, we obtain that the probabiity of reating a fase positive iue C 0 E at distane δ from one of the iues C,..., C m orresponding to the set S is upper bounded by og ρ δ = og ρ δ = og ρ ρ K δ +δ δ+δ ρ K δ +δ δ+δ ρ og δ+ ρ hδ/ K +δ δ+δ, where h denotes the binary entropy funtion.

4 Sine 0 < ρ <, to show that this ast expression goes to zero exponentiay uiky as, it suffies to show that the foowing hods: δ + δ + δ og ρ h δ/ K +δ δ. 8 [Note that the right hand side of 8 approahes infinity for any δ > 0 as.] It remains to show that 8 hods. Sine K < and sine δ, we have δ + K > δ + + δ + δ δ + δδ + δ 7 δ δ δ δ, where the ast ineuaity hods based on the foowing reasoning: if δ = Ωand δ by definition, then the two dominating terms are 7 δ > δ /, and the other two terms are sma; if δ = o, then 6 δ > sine δ and the two other terms are sma ompared with δ. Moreover, for any δ, og ρ h δ/ + δ approahes infinity as. Therefore, 8 hods, and so we have shown that P Cx 0 E x0 / S exp{ Θ} as. The proof of Theorem now foows by bounding the fase positive rate over a possibe Hamming distanes. Proof: [Proof of Theorem ] Lemma gives that the probabiity of introduing a iue orresponding to an eement x 0 at a distane dx 0, S = δ from S deays exponentiay uiky as. Taking the union bound over a distanes δ =,,..., gives that the probabiity of introduing a fase positive iue into the graph is upper bounded by exp{ Θ} = exp{ Θ}. i= IV. EXTENSION TO HYPERGRAPHS The SSAMs introdued in Setion II are based on a graph onstrution. Speifiay, eah eement x i inserted into the memory is represented as a iue in a -partite graph. In this setion we generaize this onstrution so that the set S is represented by the edge struture of a hypergraph. Speifiay, et us onsider a -reguar hypergraph H with the same set of verties V as above, and with eah hyperedge being a set of verties for some positive integer between and. Let V x = {v,wx, v,wx,..., v,wx} uster uster uster uster 4 Fig.. Iustration of the storing proess of the message in a tridimensiona hypergraph =. denote the verties impiated by the eement x U. For >, when inserting x into the assoiative memory, we add a hyperedges orresponding to subsets of verties in V x. In genera this orresponds to a hyperiue ontaining hyperedges. The ase = is euivaent to the graph onstrution desribed in Setion II. Another interesting ase is =, for whih there is a bijetion between the sets of input messages and the possibe networks. Figure shows an exampe for = 4, =, and =. Sine H is sti -partite, and there are verties in eah partition, and sine H is aso -reguar, the memory reuired to store suh a hypergraph is bits. Simiar to the graphbased SSAM, it is ear that the hypergraph SSAM H- SSAM wi not produe fase negatives. As more eements are inserted, there wi be more hyperedges in the network, and so the fase positive rate inreases. The fase positive rate for H-SSAMs an be omputed using simiar arguments to those as in Setion III. Here we sketh the main ideas. The probabiity that a given hyperedge is in H after inserting m eements is now ρ = m. Thus, for a fixed density ρ, the number of messages stored shoud grow ike m og. ρ The probabiity that a random message x 0 is aepted by the network is p e = ρ, whih is the probabiity that the orresponding hyperedges are in network after having inserted the m eements. The expeted number of messages aepted by the network is { } m e = p e = exp ogρ + og. Thus, m e = o if and ony if ogρ + og. Given and fixed 0 < ρ <, a suffiient ondition is that og K og, ρ

5 where K < is a onstant. Euivaenty, it is suffiient to have ρ K. Now, et us re-express the number of bits, L, used to represent H in memory in terms of ρ and K. Sine there are possibe hyperedges, eah of whih is either present or not in H, we have that L ρ K. Ceary, L oud be redued by using some ompression tehniue, espeiay if ρ is far from 0.5. Sine m = og ρ K ρ and n = = ρ K, it foows that m = on. It an be shown that the number of bits reuired to represent a random set S of size S = m of eements drawn from a universe of size U = n is ower bounded by the entropy of suh a set whih saes as HS n og n n m og n m L HS m og n as n. Ceary, sine n = and, we have n as. Observe that the ratio of L to HS obeys ρ K K og ρ og ρ K ρ og K ogρ og ρ as. This ratio is minimum, and hene the H-SSAMs are most effiient, when ρ = 0.5. In this ase, when K, the memory reuired by H-SSAMs is within a fator of / og.44 of the optima vaue. Fixing ρ = 0.5 and fixing, we an derive the number of messages m that gives the best effiieny. It is eua to og K. Rea that n = K. Combining these expressions gives that the orresponding H-SSAM is most effiient i.e., the number of bits reuired to represent the H-SSAM is osest to the ower bound when the number of messages stored is m = ogn og = Θn n /. V. COMPARISON WITH OTHER SET IMPLEMENTATION TECHNIQUES Let us fix ρ = 0.5, and rea that, in this ase, m = Θ and og n = Θ. Tabe ompares the ompexities of various set impementation tehniues expressed as a funtion of. In these omparisons we have assumed that eah approah is impemented on a mahine whih performs operations on up to og bits in onstant time. Rea that a Boom fiter is guaranteed to have a fase positive rate of no more than ɛ if b m og e og /ɛ bits and k = b/m og e hash funtions are used. For the given density ρ = 0.5, the fase positive rate of the H-SSAM is. Thus, to ahieve a omparabe fase positive rate whie storing the same number of messages, a Boom fiter reuires Θ bits of memory, whih is of the same order as that reuired by H-SSAMs. Hene, Boom fiters and H-SSAMs are euivaent from the perspetive of ompexity. VI. CONCLUSION This paper proposes the use of sparse strutured assoiative memories SSAMs as a mehanism for effiient representation for sets when one needs to perform set membership ueries and some errors are toerabe. The basi onstrution represents the set of eements in terms of the edge struture of a graph. We aso propose an extension where the set is enoded in the struture of a hypergraph H-SSAMs, and we show that this approah has advantages in terms of redued error rate in appropriatey arge regimes. In omparison to Boom fiters, SSAMs and H-SSAMs have the advantage that they an be used in situations where error orretion is reuired. For exampe, if the ueried eement x is represented as a string of symbos from an aphabet of size, if some symbos are erased or orrupted, Boom fiters are no onger appiabe sine the entire desription of x is needed for hashing. Assoiative memories, on the other hand, are designed to aow for reovery of orrupted or missing data, and their abiity to orret for errors is studied in [6], [7]. SSAMs and H-SSAMs may be of interest for use in distributed appiations, e.g., by storing the eements of the adjaeny matrix/tensor at different agents. In this manner, no singe agent oud independenty determine if an eement is in the set, and they must ooperate when responding to ueries. This may be of interest in distributed databases with sensitive ontent. By using hash funtions to map eements to random bit indies, Boom fiters provide universa ompression of sets in the sense that the fase positive rates hod regardess of how the eements of S are drawn. However, the anaysis in this paper is onduted under the assumption that eements are drawn uniformy and independenty. We are urrenty investigating the performane of SSAMs under more genera distributions on S. One simpe way to irumvent orreation is to append the binary representation of eah eement x with additiona bits whih are drawn uniformy and independenty; of ourse, this redues the orreation but aso inreases the amount of memory reuired. Finay, we note that assoiative memories with strutures simiar to SSAMs have reenty been shown to enjoy nie faut toerane properties [8], in the sense that the error rates are not severey impated if the graph struture is perturbed i.e., edges are randomy added or removed. Ahieving this faut toerane reuires a sighty more sophistiated, iterative ookup proedure for performing ueries. In genera there are many interesting simiarities between Boom fiters and SSAMs, and estabishing more onretey the onnetion between these two strutures is the subjet of future work.

6 TABLE I COMPARISON OF THE COMPLEXITIES OF VARIOUS SET IMPLEMENTATION TECHNIQUES. Impementation Average ompexity Worst ase ompexity Memory ompexity Insert Query Insert Query Naive array Θ Hash tabe Θ Trie Θ Sef-baaning binary searh tree Θ Boom fiter Θ H-SSAM Θ ACKNOWLEDGEMENTS This work was funded in part by the European Researh Couni ERC AdG 0 NEUCOD, the Natura Sienes and Engineering Researh Couni of Canada NSERC, and the Fonds Québéoos de a reherhe sur a nature et es tehnoogies FQRNT. REFERENCES [] A. Papadogiannakis, M. Poyhronakis, and E. P. Markatos, Improving the auray of network intrusion detetion systems under oad using seetive paket disarding, in Proeedings of the Third European Workshop on System Seurity, ser. EUROSEC 0, New York, NY, USA, 00, pp. 5. [] C. S. Lin, D. C. P. Smith, and J. M. Smith, The design of a rotating assoiative memory for reationa database appiations, ACM Trans. Database Syst., vo., pp. 5 65, Mar [] A. Rajaraman and J. Uman, Mining of Massive Datasets. Cambridge University Press, 0. [4] B. Boom, Spae/time trade-offs in hash oding with aowabe errors, Communiations of the ACM, vo., no. 7, pp. 4 46, 970. [5] A. Broder and M. Mitzenmaher, Network appiations of Boom fiters: A survey, Internet Mathematis, vo., no. 4, pp , 005. [6] V. Gripon and C. Berrou, Sparse neura networks with arge earning diversity, IEEE Transations on Neura Networks, vo., no. 7, pp , Juy 0. [7], Neary-optima assoiative memories based on distributed onstant weight odes, in Proeedings of Information Theory and Appiations Workshop, San Diego, CA, USA, February 0, pp [8] F. Ledu-Primeau, V. Gripon, M. Rabbat, and W. Gross, Faut-toerant assoiative memories based on ustered graphs, May 0, submitted.

7 Max-Flow Problems. Business Computing and Operations Research 608

7 Max-Flow Problems. Business Computing and Operations Research 608 7 Max-Flow Problems Business Computing and Operations Researh 68 7. Max-Flow Problems In what follows, we onsider a somewhat modified problem onstellation Instead of osts of transmission, vetor now indiates

More information

Nonreversibility of Multiple Unicast Networks

Nonreversibility of Multiple Unicast Networks Nonreversibility of Multiple Uniast Networks Randall Dougherty and Kenneth Zeger September 27, 2005 Abstrat We prove that for any finite direted ayli network, there exists a orresponding multiple uniast

More information

SUPPLEMENTARY INFORMATION

SUPPLEMENTARY INFORMATION DOI: 10.1038/NPHOTON.013.10 Suppementary Information: Quantum teeportation using a ight emitting diode J. Nisson 1, R. M. Stevenson 1*, K. H. A. Chan 1,, J. Skiba-Szymanska 1, M. Luamarini 1, M. B. Ward

More information

CMSC 451: Lecture 9 Greedy Approximation: Set Cover Thursday, Sep 28, 2017

CMSC 451: Lecture 9 Greedy Approximation: Set Cover Thursday, Sep 28, 2017 CMSC 451: Leture 9 Greedy Approximation: Set Cover Thursday, Sep 28, 2017 Reading: Chapt 11 of KT and Set 54 of DPV Set Cover: An important lass of optimization problems involves overing a ertain domain,

More information

Sensitivity Analysis in Markov Networks

Sensitivity Analysis in Markov Networks Sensitivity Analysis in Markov Networks Hei Chan and Adnan Darwihe Computer Siene Department University of California, Los Angeles Los Angeles, CA 90095 {hei,darwihe}@s.ula.edu Abstrat This paper explores

More information

arxiv: v2 [cond-mat.dis-nn] 29 Mar 2017

arxiv: v2 [cond-mat.dis-nn] 29 Mar 2017 arxiv:69.8375v [ond-mat.dis-nn] 9 Mar 7 The distribution of first hitting times of non-baktraking random waks on Erdős-Rényi networks Ido Tishby, Ofer Biham and Eytan Katzav Raah Institute of Physis, The

More information

FOR many years the authors of this paper have worked on

FOR many years the authors of this paper have worked on The Fast Parametri Integra Equations System for Poygona D Potentia Probems Andrzej Kużeewski and Eugeniusz Zieniuk Abstrat Appiation of tehniques for modeing of boundary vaue probems impies three onfiting

More information

Alphanumeric Character Recognition

Alphanumeric Character Recognition Amerian Journa of Signa Proessing. 0; (): 34-39 DOI: 0. 593/j.ajsp.000.06 Training Tangent Simiarities with N-SVM for Aphanumeri Charater Reognition Hassiba Nemmour *, Youef Chibani Signa Proessing Laboratory,

More information

arxiv: v1 [cs.db] 25 Jun 2013

arxiv: v1 [cs.db] 25 Jun 2013 Communication Steps for Parae Query Processing Pau Beame, Paraschos Koutris and Dan Suciu {beame,pkoutris,suciu}@cs.washington.edu University of Washington arxiv:1306.5972v1 [cs.db] 25 Jun 2013 June 26,

More information

arxiv: v1 [math.co] 17 Dec 2018

arxiv: v1 [math.co] 17 Dec 2018 On the Extrema Maximum Agreement Subtree Probem arxiv:1812.06951v1 [math.o] 17 Dec 2018 Aexey Markin Department of omputer Science, Iowa State University, USA amarkin@iastate.edu Abstract Given two phyogenetic

More information

Time-varying Stiffness Characteristics of Shaft with Slant Crack

Time-varying Stiffness Characteristics of Shaft with Slant Crack Internationa Conferene on Modeing, Simuation and Appied Mathematis (MSAM 05) Time-varying Stiffness Charateristis of Shaft with Sant Cra Hengheng Xia Shoo of Aeronautia Manufaturing Engineering Nanhang

More information

c-perfect Hashing Schemes for Binary Trees, with Applications to Parallel Memories

c-perfect Hashing Schemes for Binary Trees, with Applications to Parallel Memories -Perfet Hashing Shemes for Binary Trees, with Appliations to Parallel Memories (Extended Abstrat Gennaro Cordaso 1, Alberto Negro 1, Vittorio Sarano 1, and Arnold L.Rosenberg 2 1 Dipartimento di Informatia

More information

Danielle Maddix AA238 Final Project December 9, 2016

Danielle Maddix AA238 Final Project December 9, 2016 Struture and Parameter Learning in Bayesian Networks with Appliations to Prediting Breast Caner Tumor Malignany in a Lower Dimension Feature Spae Danielle Maddix AA238 Final Projet Deember 9, 2016 Abstrat

More information

Packing Plane Spanning Trees into a Point Set

Packing Plane Spanning Trees into a Point Set Paking Plane Spanning Trees into a Point Set Ahmad Biniaz Alfredo Garía Abstrat Let P be a set of n points in the plane in general position. We show that at least n/3 plane spanning trees an be paked into

More information

CS229 Lecture notes. Andrew Ng

CS229 Lecture notes. Andrew Ng CS229 Lecture notes Andrew Ng Part IX The EM agorithm In the previous set of notes, we taked about the EM agorithm as appied to fitting a mixture of Gaussians. In this set of notes, we give a broader view

More information

Millennium Relativity Acceleration Composition. The Relativistic Relationship between Acceleration and Uniform Motion

Millennium Relativity Acceleration Composition. The Relativistic Relationship between Acceleration and Uniform Motion Millennium Relativity Aeleration Composition he Relativisti Relationship between Aeleration and niform Motion Copyright 003 Joseph A. Rybzyk Abstrat he relativisti priniples developed throughout the six

More information

The Square of the Dirichlet-to-Neumann map equals minus Laplacian

The Square of the Dirichlet-to-Neumann map equals minus Laplacian The Square of the Dirihet-to-Neumann map equas minus Lapaian D V Ingerman Abstrat. The Dirihet-to-Neumann maps onnet boundary vaues of harmoni funtions. It is an amazing fat that the square of the non-oa

More information

Lightpath routing for maximum reliability in optical mesh networks

Lightpath routing for maximum reliability in optical mesh networks Vol. 7, No. 5 / May 2008 / JOURNAL OF OPTICAL NETWORKING 449 Lightpath routing for maximum reliability in optial mesh networks Shengli Yuan, 1, * Saket Varma, 2 and Jason P. Jue 2 1 Department of Computer

More information

Robust Recovery of Signals From a Structured Union of Subspaces

Robust Recovery of Signals From a Structured Union of Subspaces Robust Reovery of Signals From a Strutured Union of Subspaes 1 Yonina C. Eldar, Senior Member, IEEE and Moshe Mishali, Student Member, IEEE arxiv:87.4581v2 [nlin.cg] 3 Mar 29 Abstrat Traditional sampling

More information

RECENTLY, extremely low error probability has been required. New Decoding Scheme for LDPC Codes Based on Simple Product Code Structure

RECENTLY, extremely low error probability has been required. New Decoding Scheme for LDPC Codes Based on Simple Product Code Structure JOURNAL OF COMMUNICATIONS AND NETWORKS, VOL 17, NO 4, AUGUST 2015 351 New Deoding Sheme for LDPC Codes Based on Simpe Produt Code Struture Beomkyu Shin, Seokbeom Hong, Hosung Park, Jong-Seon No, and Dong-Joon

More information

Maximum Entropy and Exponential Families

Maximum Entropy and Exponential Families Maximum Entropy and Exponential Families April 9, 209 Abstrat The goal of this note is to derive the exponential form of probability distribution from more basi onsiderations, in partiular Entropy. It

More information

Complexity of Regularization RBF Networks

Complexity of Regularization RBF Networks Complexity of Regularization RBF Networks Mark A Kon Department of Mathematis and Statistis Boston University Boston, MA 02215 mkon@buedu Leszek Plaskota Institute of Applied Mathematis University of Warsaw

More information

ABSOLUTELY CONTINUOUS FUNCTIONS OF TWO VARIABLES IN THE SENSE OF CARATHÉODORY

ABSOLUTELY CONTINUOUS FUNCTIONS OF TWO VARIABLES IN THE SENSE OF CARATHÉODORY Eetroni Journa of Differentia Equations, Vo. 2010(2010), No. 154, pp. 1 11. ISSN: 1072-6691. URL: http://ejde.math.txstate.edu or http://ejde.math.unt.edu ftp ejde.math.txstate.edu ABSOLUTELY CONTINUOUS

More information

Steepest Descent Adaptation of Min-Max Fuzzy If-Then Rules 1

Steepest Descent Adaptation of Min-Max Fuzzy If-Then Rules 1 Steepest Descent Adaptation of Min-Max Fuzzy If-Then Rues 1 R.J. Marks II, S. Oh, P. Arabshahi Λ, T.P. Caude, J.J. Choi, B.G. Song Λ Λ Dept. of Eectrica Engineering Boeing Computer Services University

More information

Genetic Algorithm Based Recurrent Fuzzy Neural Network Modeling of Chemical Processes

Genetic Algorithm Based Recurrent Fuzzy Neural Network Modeling of Chemical Processes Journa of Universa Computer Siene, vo. 3, no. 9 (27), 332-343 submitted: 2/6/6, aepted: 24//6, appeared: 28/9/7 J.UCS Geneti Agorithm Based Reurrent Fuzzy Neura Network odeing of Chemia Proesses Jii Tao,

More information

Multi-version Coding for Consistent Distributed Storage of Correlated Data Updates

Multi-version Coding for Consistent Distributed Storage of Correlated Data Updates Multi-version Coding for Consistent Distributed Storage of Correlated Data Updates Ramy E. Ali and Vivek R. Cadambe 1 arxiv:1708.06042v1 [s.it] 21 Aug 2017 Abstrat Motivated by appliations of distributed

More information

A Brief Introduction to Markov Chains and Hidden Markov Models

A Brief Introduction to Markov Chains and Hidden Markov Models A Brief Introduction to Markov Chains and Hidden Markov Modes Aen B MacKenzie Notes for December 1, 3, &8, 2015 Discrete-Time Markov Chains You may reca that when we first introduced random processes,

More information

MARKOV CHAINS AND MARKOV DECISION THEORY. Contents

MARKOV CHAINS AND MARKOV DECISION THEORY. Contents MARKOV CHAINS AND MARKOV DECISION THEORY ARINDRIMA DATTA Abstract. In this paper, we begin with a forma introduction to probabiity and expain the concept of random variabes and stochastic processes. After

More information

The Effectiveness of the Linear Hull Effect

The Effectiveness of the Linear Hull Effect The Effetiveness of the Linear Hull Effet S. Murphy Tehnial Report RHUL MA 009 9 6 Otober 009 Department of Mathematis Royal Holloway, University of London Egham, Surrey TW0 0EX, England http://www.rhul.a.uk/mathematis/tehreports

More information

Research Article Some Applications of Second-Order Differential Subordination on a Class of Analytic Functions Defined by Komatu Integral Operator

Research Article Some Applications of Second-Order Differential Subordination on a Class of Analytic Functions Defined by Komatu Integral Operator ISRN Mathematia Anaysis, Artie ID 66235, 5 pages http://dx.doi.org/1.1155/214/66235 Researh Artie Some Appiations of Seond-Order Differentia Subordination on a Cass of Anayti Funtions Defined by Komatu

More information

A NETWORK SIMPLEX ALGORITHM FOR THE MINIMUM COST-BENEFIT NETWORK FLOW PROBLEM

A NETWORK SIMPLEX ALGORITHM FOR THE MINIMUM COST-BENEFIT NETWORK FLOW PROBLEM NETWORK SIMPLEX LGORITHM FOR THE MINIMUM COST-BENEFIT NETWORK FLOW PROBLEM Cen Çalışan, Utah Valley University, 800 W. University Parway, Orem, UT 84058, 801-863-6487, en.alisan@uvu.edu BSTRCT The minimum

More information

Cryptanalysis of PKP: A New Approach

Cryptanalysis of PKP: A New Approach Cryptanaysis of PKP: A New Approach Éiane Jaumes and Antoine Joux DCSSI 18, rue du Dr. Zamenhoff F-92131 Issy-es-Mx Cedex France eiane.jaumes@wanadoo.fr Antoine.Joux@ens.fr Abstract. Quite recenty, in

More information

Hankel Optimal Model Order Reduction 1

Hankel Optimal Model Order Reduction 1 Massahusetts Institute of Tehnology Department of Eletrial Engineering and Computer Siene 6.245: MULTIVARIABLE CONTROL SYSTEMS by A. Megretski Hankel Optimal Model Order Redution 1 This leture overs both

More information

BER Performance Comparison of Overlap FDE and Sliding-window Time-domain Equalization for Single-carrier Transmission

BER Performance Comparison of Overlap FDE and Sliding-window Time-domain Equalization for Single-carrier Transmission BER Performane Comparison of Overap FDE and Siding-window ime-domain Equaization for Singe-arrier ransmission atsunori OBARA Kazuki AKEDA and Fumiyuki ADACI Dept. of Eetria and Communiation Engineering,

More information

Naïve Bayes for Text Classification

Naïve Bayes for Text Classification Naïve Bayes for Text Cassifiation adapted by Lye Ungar from sides by Mith Marus, whih were adapted from sides by Massimo Poesio, whih were adapted from sides by Chris Manning : Exampe: Is this spam? From:

More information

Assessing the Performance of a BCI: A Task-Oriented Approach

Assessing the Performance of a BCI: A Task-Oriented Approach Assessing the Performane of a BCI: A Task-Oriented Approah B. Dal Seno, L. Mainardi 2, M. Matteui Department of Eletronis and Information, IIT-Unit, Politenio di Milano, Italy 2 Department of Bioengineering,

More information

Math 124B January 31, 2012

Math 124B January 31, 2012 Math 124B January 31, 212 Viktor Grigoryan 7 Inhomogeneous boundary vaue probems Having studied the theory of Fourier series, with which we successfuy soved boundary vaue probems for the homogeneous heat

More information

Flexible Word Design and Graph Labeling

Flexible Word Design and Graph Labeling Flexible Word Design and Graph Labeling Ming-Yang Kao Manan Sanghi Robert Shweller Abstrat Motivated by emerging appliations for DNA ode word design, we onsider a generalization of the ode word design

More information

Generalized Dimensional Analysis

Generalized Dimensional Analysis #HUTP-92/A036 7/92 Generalized Dimensional Analysis arxiv:hep-ph/9207278v1 31 Jul 1992 Howard Georgi Lyman Laboratory of Physis Harvard University Cambridge, MA 02138 Abstrat I desribe a version of so-alled

More information

7.1 Roots of a Polynomial

7.1 Roots of a Polynomial 7.1 Roots of a Polynomial A. Purpose Given the oeffiients a i of a polynomial of degree n = NDEG > 0, a 1 z n + a 2 z n 1 +... + a n z + a n+1 with a 1 0, this subroutine omputes the NDEG roots of the

More information

Efficiently Generating Random Bits from Finite State Markov Chains

Efficiently Generating Random Bits from Finite State Markov Chains 1 Efficienty Generating Random Bits from Finite State Markov Chains Hongchao Zhou and Jehoshua Bruck, Feow, IEEE Abstract The probem of random number generation from an uncorreated random source (of unknown

More information

The Hanging Chain. John McCuan. January 19, 2006

The Hanging Chain. John McCuan. January 19, 2006 The Hanging Chain John MCuan January 19, 2006 1 Introdution We onsider a hain of length L attahed to two points (a, u a and (b, u b in the plane. It is assumed that the hain hangs in the plane under a

More information

Value-aware Resource Allocation for Service Guarantees in Networks

Value-aware Resource Allocation for Service Guarantees in Networks Vaue-aware Resoure Aoation for Servie Guarantees in Networks Parima Parag, Srinivas Shakkottai and Jean-François Chamberand Department of Eetria and Computer Engineering Texas A&M University, Coege Station,

More information

Uniprocessor Feasibility of Sporadic Tasks with Constrained Deadlines is Strongly conp-complete

Uniprocessor Feasibility of Sporadic Tasks with Constrained Deadlines is Strongly conp-complete Uniprocessor Feasibiity of Sporadic Tasks with Constrained Deadines is Strongy conp-compete Pontus Ekberg and Wang Yi Uppsaa University, Sweden Emai: {pontus.ekberg yi}@it.uu.se Abstract Deciding the feasibiity

More information

An Algorithm for Pruning Redundant Modules in Min-Max Modular Network

An Algorithm for Pruning Redundant Modules in Min-Max Modular Network An Agorithm for Pruning Redundant Modues in Min-Max Moduar Network Hui-Cheng Lian and Bao-Liang Lu Department of Computer Science and Engineering, Shanghai Jiao Tong University 1954 Hua Shan Rd., Shanghai

More information

Moment - Axial Force Interaction Report and Program

Moment - Axial Force Interaction Report and Program oment - Aia Fore Interation Report and rogram ehdi Vojoudi Vojoudi@vojoudi.om THINK GLOBAL, ACT LOCAL Spring 3 Strutura Stabiity rojet. Sope Cauate the moment aia fore interation urve for a hinged oumn

More information

Optimization of Statistical Decisions for Age Replacement Problems via a New Pivotal Quantity Averaging Approach

Optimization of Statistical Decisions for Age Replacement Problems via a New Pivotal Quantity Averaging Approach Amerian Journal of heoretial and Applied tatistis 6; 5(-): -8 Published online January 7, 6 (http://www.sienepublishinggroup.om/j/ajtas) doi:.648/j.ajtas.s.65.4 IN: 36-8999 (Print); IN: 36-96 (Online)

More information

ONLINE APPENDICES for Cost-Effective Quality Assurance in Crowd Labeling

ONLINE APPENDICES for Cost-Effective Quality Assurance in Crowd Labeling ONLINE APPENDICES for Cost-Effetive Quality Assurane in Crowd Labeling Jing Wang Shool of Business and Management Hong Kong University of Siene and Tehnology Clear Water Bay Kowloon Hong Kong jwang@usthk

More information

Efficient Generation of Random Bits from Finite State Markov Chains

Efficient Generation of Random Bits from Finite State Markov Chains Efficient Generation of Random Bits from Finite State Markov Chains Hongchao Zhou and Jehoshua Bruck, Feow, IEEE Abstract The probem of random number generation from an uncorreated random source (of unknown

More information

Batch Process Monitoring Using Two-Dimensional Hidden Semi-Markov Models

Batch Process Monitoring Using Two-Dimensional Hidden Semi-Markov Models Bath Proess Monitoring Using wo-imensiona Hidden Semi-Marov Modes J. Chen, Member, IAENG and Y.-C. Jiang Abstrat In this paper, a nove monitoring method for the repetitive bath operation with two-dimensiona

More information

Measuring & Inducing Neural Activity Using Extracellular Fields I: Inverse systems approach

Measuring & Inducing Neural Activity Using Extracellular Fields I: Inverse systems approach Measuring & Induing Neural Ativity Using Extraellular Fields I: Inverse systems approah Keith Dillon Department of Eletrial and Computer Engineering University of California San Diego 9500 Gilman Dr. La

More information

LECTURE NOTES FOR , FALL 2004

LECTURE NOTES FOR , FALL 2004 LECTURE NOTES FOR 18.155, FALL 2004 83 12. Cone support and wavefront set In disussing the singular support of a tempered distibution above, notie that singsupp(u) = only implies that u C (R n ), not as

More information

Coding for Random Projections and Approximate Near Neighbor Search

Coding for Random Projections and Approximate Near Neighbor Search Coding for Random Projetions and Approximate Near Neighbor Searh Ping Li Department of Statistis & Biostatistis Department of Computer Siene Rutgers University Pisataay, NJ 8854, USA pingli@stat.rutgers.edu

More information

A simple expression for radial distribution functions of pure fluids and mixtures

A simple expression for radial distribution functions of pure fluids and mixtures A simple expression for radial distribution funtions of pure fluids and mixtures Enrio Matteoli a) Istituto di Chimia Quantistia ed Energetia Moleolare, CNR, Via Risorgimento, 35, 56126 Pisa, Italy G.

More information

THE THREE POINT STEINER PROBLEM ON THE FLAT TORUS: THE MINIMAL LUNE CASE

THE THREE POINT STEINER PROBLEM ON THE FLAT TORUS: THE MINIMAL LUNE CASE THE THREE POINT STEINER PROBLEM ON THE FLAT TORUS: THE MINIMAL LUNE CASE KATIE L. MAY AND MELISSA A. MITCHELL Abstract. We show how to identify the minima path network connecting three fixed points on

More information

The research of Q-ary LDPC codes in the navigation system

The research of Q-ary LDPC codes in the navigation system 3rd Internationa Conferene on Mehatronis and Information Tehnoogy (ICMIT 26) The researh of Q-ary LDPC odes in the naigation system Xu Yi, Daoxing Guo2 and Wei You3 Coege of Communiations Engineering,

More information

Available online at ScienceDirect. Energy Procedia 76 (2015 ) European Geosciences Union General Assembly 2015, EGU

Available online at  ScienceDirect. Energy Procedia 76 (2015 ) European Geosciences Union General Assembly 2015, EGU Avaiabe onine at www.sienediret.om SieneDiret Energy Proedia 76 (05 ) 406 4 European Geosienes Union Genera Assemby 05, EGU Division Energy, Resoures & Environment, ERE Appiation of stoasti metods to doube

More information

An iterative least-square method suitable for solving large sparse matrices

An iterative least-square method suitable for solving large sparse matrices An iteratie least-square method suitable for soling large sparse matries By I. M. Khabaza The purpose of this paper is to report on the results of numerial experiments with an iteratie least-square method

More information

Figure 2 As explained previously [1] the relative length L v of the train in motion with respect to the train at rest L s, is: =

Figure 2 As explained previously [1] the relative length L v of the train in motion with respect to the train at rest L s, is: = "Expaining the Iusion of the Constant Veoity of Light." Pau Marmet 401 Ogivie Road, Gouester, Ontario, Canada K1J 7N4 E-Mai: Pau.Marmet@Ottawa.om ABSTRACT. Considering that photons trave at the veoity

More information

Iterative Decoding Performance Bounds for LDPC Codes on Noisy Channels

Iterative Decoding Performance Bounds for LDPC Codes on Noisy Channels Iterative Decoding Performance Bounds for LDPC Codes on Noisy Channes arxiv:cs/060700v1 [cs.it] 6 Ju 006 Chun-Hao Hsu and Achieas Anastasopouos Eectrica Engineering and Computer Science Department University

More information

XSAT of linear CNF formulas

XSAT of linear CNF formulas XSAT of inear CN formuas Bernd R. Schuh Dr. Bernd Schuh, D-50968 Kön, Germany; bernd.schuh@netcoogne.de eywords: compexity, XSAT, exact inear formua, -reguarity, -uniformity, NPcompeteness Abstract. Open

More information

14.471: Fall 2012:PS3 Solutions

14.471: Fall 2012:PS3 Solutions 4.47: Fa 202:PS3 Soutions Deember 4, 202. Commodity Taxation Consider an eonomy onsisting of identia individuas with preferenes given by U (x,x 2,)= α og (x )+ α 2 og (x 2 )+( α α 2 ) og ( ). Individuas

More information

THE REACHABILITY CONES OF ESSENTIALLY NONNEGATIVE MATRICES

THE REACHABILITY CONES OF ESSENTIALLY NONNEGATIVE MATRICES THE REACHABILITY CONES OF ESSENTIALLY NONNEGATIVE MATRICES by Michae Neumann Department of Mathematics, University of Connecticut, Storrs, CT 06269 3009 and Ronad J. Stern Department of Mathematics, Concordia

More information

Computer Science 786S - Statistical Methods in Natural Language Processing and Data Analysis Page 1

Computer Science 786S - Statistical Methods in Natural Language Processing and Data Analysis Page 1 Computer Siene 786S - Statistial Methods in Natural Language Proessing and Data Analysis Page 1 Hypothesis Testing A statistial hypothesis is a statement about the nature of the distribution of a random

More information

hole h vs. e configurations: l l for N > 2 l + 1 J = H as example of localization, delocalization, tunneling ikx k

hole h vs. e configurations: l l for N > 2 l + 1 J = H as example of localization, delocalization, tunneling ikx k Infinite 1-D Lattice CTDL, pages 1156-1168 37-1 LAST TIME: ( ) ( ) + N + 1 N hoe h vs. e configurations: for N > + 1 e rij unchanged ζ( NLS) ζ( NLS) [ ζn unchanged ] Hund s 3rd Rue (Lowest L - S term of

More information

Recursive Constructions of Parallel FIFO and LIFO Queues with Switched Delay Lines

Recursive Constructions of Parallel FIFO and LIFO Queues with Switched Delay Lines Recursive Constructions of Parae FIFO and LIFO Queues with Switched Deay Lines Po-Kai Huang, Cheng-Shang Chang, Feow, IEEE, Jay Cheng, Member, IEEE, and Duan-Shin Lee, Senior Member, IEEE Abstract One

More information

After the completion of this section the student should recall

After the completion of this section the student should recall Chapter I MTH FUNDMENTLS I. Sets, Numbers, Coordinates, Funtions ugust 30, 08 3 I. SETS, NUMERS, COORDINTES, FUNCTIONS Objetives: fter the ompletion of this setion the student should reall - the definition

More information

Expectation-Maximization for Estimating Parameters for a Mixture of Poissons

Expectation-Maximization for Estimating Parameters for a Mixture of Poissons Expectation-Maximization for Estimating Parameters for a Mixture of Poissons Brandon Maone Department of Computer Science University of Hesini February 18, 2014 Abstract This document derives, in excrutiating

More information

8 APPENDIX. E[m M] = (n S )(1 exp( exp(s min + c M))) (19) E[m M] n exp(s min + c M) (20) 8.1 EMPIRICAL EVALUATION OF SAMPLING

8 APPENDIX. E[m M] = (n S )(1 exp( exp(s min + c M))) (19) E[m M] n exp(s min + c M) (20) 8.1 EMPIRICAL EVALUATION OF SAMPLING 8 APPENDIX 8.1 EMPIRICAL EVALUATION OF SAMPLING We wish to evauate the empirica accuracy of our samping technique on concrete exampes. We do this in two ways. First, we can sort the eements by probabiity

More information

Searching All Approximate Covers and Their Distance using Finite Automata

Searching All Approximate Covers and Their Distance using Finite Automata Searhing All Approximate Covers and Their Distane using Finite Automata Ondřej Guth, Bořivoj Melihar, and Miroslav Balík České vysoké učení tehniké v Praze, Praha, CZ, {gutho1,melihar,alikm}@fel.vut.z

More information

A. Distribution of the test statistic

A. Distribution of the test statistic A. Distribution of the test statistic In the sequentia test, we first compute the test statistic from a mini-batch of size m. If a decision cannot be made with this statistic, we keep increasing the mini-batch

More information

Counting Arbitrary Subgraphs in Data Streams

Counting Arbitrary Subgraphs in Data Streams Counting Arbitrary Subgraphs in Data Streams Daniel M. Kane 1, Kurt Mehlhorn, Thomas Sauerwald, and He Sun 1 Department of Mathematis, Stanford University, USA Max Plank Institute for Informatis, Germany

More information

Sequence Analysis, WS 14/15, D. Huson & R. Neher (this part by D. Huson & J. Fischer) January 21,

Sequence Analysis, WS 14/15, D. Huson & R. Neher (this part by D. Huson & J. Fischer) January 21, Sequene Analysis, WS 14/15, D. Huson & R. Neher (this part by D. Huson & J. Fisher) January 21, 201511 9 Suffix Trees and Suffix Arrays This leture is based on the following soures, whih are all reommended

More information

How many random edges make a dense hypergraph non-2-colorable?

How many random edges make a dense hypergraph non-2-colorable? How many random edges make a dense hypergraph non--coorabe? Benny Sudakov Jan Vondrák Abstract We study a mode of random uniform hypergraphs, where a random instance is obtained by adding random edges

More information

Time Domain Method of Moments

Time Domain Method of Moments Time Domain Method of Moments Massahusetts Institute of Tehnology 6.635 leture notes 1 Introdution The Method of Moments (MoM) introdued in the previous leture is widely used for solving integral equations

More information

MINIMIZATION OF FREQUENCY-WEIGHTED l 2 -SENSITIVITY FOR MULTI-INPUT/MULTI-OUTPUT LINEAR SYSTEMS

MINIMIZATION OF FREQUENCY-WEIGHTED l 2 -SENSITIVITY FOR MULTI-INPUT/MULTI-OUTPUT LINEAR SYSTEMS Automatique et ordinateurs MINIMIZAION OF FREQUENCY-WEIGHED -SENSIIVIY FOR MULI-INPU/MULI-OUPU LINEAR SYSEMS AKAO HINAMOO, OSAMU ANAKA, AKIMISU DOI Key words: MIMO inear disrete-time systems, Frequeny-weighted

More information

Using the Green s Function to find the Solution to the Wave. Equation:

Using the Green s Function to find the Solution to the Wave. Equation: Using the Green s Funtion to find the Soution to the Wave Exampe 1: 2 1 2 2 t 2 Equation: r,t q 0 e it r aẑ r aẑ r,t r 1 r ; r r,t r 1 r 2 The Green s funtion soution is given by r,t G R r r,t t Fr,t d

More information

REFINED UPPER BOUNDS FOR THE LINEAR DIOPHANTINE PROBLEM OF FROBENIUS. 1. Introduction

REFINED UPPER BOUNDS FOR THE LINEAR DIOPHANTINE PROBLEM OF FROBENIUS. 1. Introduction Version of 5/2/2003 To appear in Advanes in Applied Mathematis REFINED UPPER BOUNDS FOR THE LINEAR DIOPHANTINE PROBLEM OF FROBENIUS MATTHIAS BECK AND SHELEMYAHU ZACKS Abstrat We study the Frobenius problem:

More information

max min z i i=1 x j k s.t. j=1 x j j:i T j

max min z i i=1 x j k s.t. j=1 x j j:i T j AM 221: Advaned Optimization Spring 2016 Prof. Yaron Singer Leture 22 April 18th 1 Overview In this leture, we will study the pipage rounding tehnique whih is a deterministi rounding proedure that an be

More information

Counting Idempotent Relations

Counting Idempotent Relations Counting Idempotent Relations Beriht-Nr. 2008-15 Florian Kammüller ISSN 1436-9915 2 Abstrat This artile introdues and motivates idempotent relations. It summarizes haraterizations of idempotents and their

More information

Singular Event Detection

Singular Event Detection Singular Event Detetion Rafael S. Garía Eletrial Engineering University of Puerto Rio at Mayagüez Rafael.Garia@ee.uprm.edu Faulty Mentor: S. Shankar Sastry Researh Supervisor: Jonathan Sprinkle Graduate

More information

On the Complexity of the Weighted Fused Lasso

On the Complexity of the Weighted Fused Lasso ON THE COMPLEXITY OF THE WEIGHTED FUSED LASSO On the Compleity of the Weighted Fused Lasso José Bento jose.bento@b.edu Ralph Furmaniak rf@am.org Surjyendu Ray rays@b.edu Abstrat The solution path of the

More information

PHYS 110B - HW #1 Fall 2005, Solutions by David Pace Equations referenced as Eq. # are from Griffiths Problem statements are paraphrased

PHYS 110B - HW #1 Fall 2005, Solutions by David Pace Equations referenced as Eq. # are from Griffiths Problem statements are paraphrased PHYS 110B - HW #1 Fa 2005, Soutions by David Pace Equations referenced as Eq. # are from Griffiths Probem statements are paraphrased [1.] Probem 6.8 from Griffiths A ong cyinder has radius R and a magnetization

More information

Parallel disrete-event simulation is an attempt to speed-up the simulation proess through the use of multiple proessors. In some sense parallel disret

Parallel disrete-event simulation is an attempt to speed-up the simulation proess through the use of multiple proessors. In some sense parallel disret Exploiting intra-objet dependenies in parallel simulation Franeso Quaglia a;1 Roberto Baldoni a;2 a Dipartimento di Informatia e Sistemistia Universita \La Sapienza" Via Salaria 113, 198 Roma, Italy Abstrat

More information

New Algorithms for Nonlinear Generalized Disjunctive Programming

New Algorithms for Nonlinear Generalized Disjunctive Programming ew Agorithms for oninear Generaized Disuntive Programming Sangbum Lee and Ignaio E. Grossmann * Department of Chemia Engineering Carnegie Meon University Pittsburgh PA U.S.A. Otober 999 / Marh * To whom

More information

FRIEZE GROUPS IN R 2

FRIEZE GROUPS IN R 2 FRIEZE GROUPS IN R 2 MAXWELL STOLARSKI Abstract. Focusing on the Eucidean pane under the Pythagorean Metric, our goa is to cassify the frieze groups, discrete subgroups of the set of isometries of the

More information

Dept. of Computer Science. Raleigh, NC 27695, USA. May 14, Abstract. 1, u 2 q i+1 :

Dept. of Computer Science. Raleigh, NC 27695, USA. May 14, Abstract. 1, u 2 q i+1 : Anti-Leture Hall Compositions Sylvie Corteel CNRS PRiSM, UVSQ 45 Avenue des Etats-Unis 78035 Versailles, Frane syl@prism.uvsq.fr Carla D. Savage Dept. of Computer Siene N. C. State University, Box 8206

More information

State Diagrams. Margaret M. Fleck. 14 November 2011

State Diagrams. Margaret M. Fleck. 14 November 2011 State Diagrams Margaret M. Flek 14 November 2011 These notes over state diagrams. 1 Introdution State diagrams are a type of direted graph, in whih the graph nodes represent states and labels on the graph

More information

Keywords: Controllers with PI and PID dynamics, Takagi-Sugeno fuzzy models, Takagi- Sugeno fuzzy controllers, stability analysis, winding system.

Keywords: Controllers with PI and PID dynamics, Takagi-Sugeno fuzzy models, Takagi- Sugeno fuzzy controllers, stability analysis, winding system. Ata Poytehnia Hungaria Vo. 2, No., 2005 Deveopment of Conventiona and Fuzzy Controers and Takagi-Sugeno Fuzzy Modes Dediated for Contro of Low Order Benhmarks with Time Variabe Parameters Stefan Preit

More information

On colorings of the Boolean lattice avoiding a rainbow copy of a poset arxiv: v1 [math.co] 21 Dec 2018

On colorings of the Boolean lattice avoiding a rainbow copy of a poset arxiv: v1 [math.co] 21 Dec 2018 On coorings of the Booean attice avoiding a rainbow copy of a poset arxiv:1812.09058v1 [math.co] 21 Dec 2018 Baázs Patkós Afréd Rényi Institute of Mathematics, Hungarian Academy of Scinces H-1053, Budapest,

More information

IMPEDANCE EFFECTS OF LEFT TURNERS FROM THE MAJOR STREET AT A TWSC INTERSECTION

IMPEDANCE EFFECTS OF LEFT TURNERS FROM THE MAJOR STREET AT A TWSC INTERSECTION 09-1289 Citation: Brilon, W. (2009): Impedane Effets of Left Turners from the Major Street at A TWSC Intersetion. Transportation Researh Reord Nr. 2130, pp. 2-8 IMPEDANCE EFFECTS OF LEFT TURNERS FROM THE

More information

Geography, Technology, and Optimal Trade Policy with Heterogeneous Firms

Geography, Technology, and Optimal Trade Policy with Heterogeneous Firms Geography, Tehnoogy, and Optima Trade Poiy with Heterogeneous Firms Antonea Noo, Università de Saento Gianmaro I.P. Ottaviano, LSE, University of Boogna, CEP and CEPR atteo Sato, European Commission This

More information

Gauss Law. 2. Gauss s Law: connects charge and field 3. Applications of Gauss s Law

Gauss Law. 2. Gauss s Law: connects charge and field 3. Applications of Gauss s Law Gauss Law 1. Review on 1) Couomb s Law (charge and force) 2) Eectric Fied (fied and force) 2. Gauss s Law: connects charge and fied 3. Appications of Gauss s Law Couomb s Law and Eectric Fied Couomb s

More information

Stochastic Variational Inference with Gradient Linearization

Stochastic Variational Inference with Gradient Linearization Stochastic Variationa Inference with Gradient Linearization Suppementa Materia Tobias Pötz * Anne S Wannenwetsch Stefan Roth Department of Computer Science, TU Darmstadt Preface In this suppementa materia,

More information

Coding for Noisy Write-Efficient Memories

Coding for Noisy Write-Efficient Memories Coding for oisy Write-Effiient Memories Qing Li Computer Si. & Eng. Dept. Texas A & M University College Station, TX 77843 qingli@se.tamu.edu Anxiao (Andrew) Jiang CSE and ECE Departments Texas A & M University

More information

(q) -convergence. Comenius University, Bratislava, Slovakia

(q) -convergence.   Comenius University, Bratislava, Slovakia Annales Mathematiae et Informatiae 38 (2011) pp. 27 36 http://ami.ektf.hu On I (q) -onvergene J. Gogola a, M. Mačaj b, T. Visnyai b a University of Eonomis, Bratislava, Slovakia e-mail: gogola@euba.sk

More information

Sensor management for PRF selection in the track-before-detect context

Sensor management for PRF selection in the track-before-detect context Sensor management for PRF seletion in the tra-before-detet ontext Fotios Katsilieris, Yvo Boers, and Hans Driessen Thales Nederland B.V. Haasbergerstraat 49, 7554 PA Hengelo, the Netherlands Email: {Fotios.Katsilieris,

More information

Separation of Variables and a Spherical Shell with Surface Charge

Separation of Variables and a Spherical Shell with Surface Charge Separation of Variabes and a Spherica She with Surface Charge In cass we worked out the eectrostatic potentia due to a spherica she of radius R with a surface charge density σθ = σ cos θ. This cacuation

More information

COMBINED PROBE FOR MACH NUMBER, TEMPERATURE AND INCIDENCE INDICATION

COMBINED PROBE FOR MACH NUMBER, TEMPERATURE AND INCIDENCE INDICATION 4 TH INTERNATIONAL CONGRESS OF THE AERONAUTICAL SCIENCES COMBINED PROBE FOR MACH NUMBER, TEMPERATURE AND INCIDENCE INDICATION Jiri Nozika*, Josef Adame*, Daniel Hanus** *Department of Fluid Dynamis and

More information

A Characterization of Wavelet Convergence in Sobolev Spaces

A Characterization of Wavelet Convergence in Sobolev Spaces A Charaterization of Wavelet Convergene in Sobolev Spaes Mark A. Kon 1 oston University Louise Arakelian Raphael Howard University Dediated to Prof. Robert Carroll on the oasion of his 70th birthday. Abstrat

More information