Factoring Synchronous Grammars By Sorting

Size: px
Start display at page:

Download "Factoring Synchronous Grammars By Sorting"

Transcription

1 Factoring Syncronous Grammars By Sorting Daniel Gildea Computer Science Dept. Uniersity of Rocester Rocester, NY Giorgio Satta Dept. of Information Eng g Uniersity of Padua I- Padua, Italy Hao Zang Computer Science Dept. Uniersity of Rocester Rocester, NY Abstract Syncronous Context-Free Grammars (SCFGs) ae been successfully exploited as translation models in macine translation applications. Wen parsing wit an SCFG, computational complexity grows exponentially wit te lengt of te rules, in te worst case. In tis paper we examine te problem of factorizing eac rule of an input SCFG to a generatiely equialent set of rules, eac aing te smallest possible lengt. Our algoritm works in time O(n log n), for eac rule of lengt n. Tis improes upon preious results and soles an open problem about recognizing permutations tat can be factored. Introduction Syncronous Context-Free Grammars (SCFGs) are a generalization of te Context-Free Grammar (CFG) formalism to simultaneously produce strings in two languages. SCFGs ae a wide range of applications, including macine translation, word and prase alignments, and automatic dictionary construction. Variations of SCFGs go back to Ao and Ullman (9) s Syntax-Directed Translation Scemata, but also include te Inersion Transduction Grammars in Wu (99), wic restrict grammar rules to be binary, te syncronous grammars in Ciang (00), wic use only a single nonterminal symbol, and te Multitext Grammars in Melamed (00), wic allow independent rewriting, as well as oter tree-based models suc as Yamada and Knigt (00) and Galley et al. (00). Wen iewed as a rewriting system, an SCFG generates a set of string pairs, representing some translation relation. We are concerned ere wit te time complexity of parsing suc a pair, according to te grammar. Assume ten a pair wit eac string aing a maximum lengt of N, and consider an SCFG G of size G, wit a bound of n nonterminals in te rigt-and side of eac rule in a single dimension, wic we call below te rank of G. As an upper bound, parsing can be carried out in time O( G N n+ ) by a dynamic programming algoritm maintaining continuous spans in one dimension. As a lower bound, parsing strategies wit discontinuous spans in bot dimensions can take time Ω( G N c n ) for unfriendly permutations (Satta and Peserico, 00). A natural question to ask ten is: Wat if we could reduce te rank of G, presering te generated translation? As in te case of CFGs, one way of doing tis would be to factorize eac single rule into seeral rules of rank strictly smaller tan n. It is not difficult to see tat tis would result in a new grammar of size at most G. In te time complexities reported aboe, we see tat suc a size increase would be more tan compensated by te reduction in te degree of te polynomial in N. We tus conclude tat a reduction in te rank of an SCFG would result in more efficient parsing algoritms, for most common parsing strategies. In te general case, normal forms wit bounded rank are not admitted by SCFGs, as sown in (Ao and Ullman, 9). Noneteless, an SCFG wit a rank of n may not necessarily meet te worst case of Ao and Ullman (9). It is ten reasonable to ask if our SCFG G can be factorized, and wat is te smallest rank k < n tat can be obtained in tis way. Tis paper answers tese two questions, by proiding an algoritm tat factorizes te rules of an input SCFG, resulting in a new, generatiely equialent, SCFG wit rank k as low as possible. Te algoritm works in time O(n log n) for eac rule, regardless of te rank k of te factorized rules. As discussed aboe, in tis way we aciee an improement of te parsing time for SCFGs, obtaining an upper bound of O( G N k+ ) by using a parsing strategy tat maintains continuous

2 ,,,,,,,,,,,,,, Figure : Two permutation trees. Te permutations associated wit te leaes can be produced by composing te permutations at te internal nodes. spans in one dimension. Preious work on tis problem as been presented in Zang et al. (00), were a metod is proided for casting an SCFG to a form wit rank k =. If generalized to any alue of k, tat algoritm would run in time O(n ). We tus improe existing factorization metods by almost a factor of n. We also sole an open problem mentioned by Albert et al. (00), wo pose te question of weter irreducible, or simple, permutations can be recognized in time less tan Θ(n ). Syncronous CFGs and permutation trees We begin by describing te syncronous CFG formalism, wic is more rigorously defined by Ao and Ullman (9) and Satta and Peserico (00). Let us consider strings defined oer some set of nonterminal and terminal symbols, as defined for CFGs. We say tat two suc strings are syncronous if some bijectie relation is gien between te occurrences of te nonterminals in te two strings. A syncronous context-free grammar (SCFG) is defined as a CFG, wit te difference tat it uses syncronous rules of te form [A α, A α ], wit A, A nonterminals and α, α syncronous strings. We can use production [A α, A α ] to rewrite any syncronous strings [γ A γ, γ A γ ] into te syncronous strings [γ α γ, γ α γ ], under te condition tat te indicated occurrences of A and A be related by te bijection associated wit te source syncronous strings. Furtermore, te bijectie relation associated wit te target syncronous strings is obtained by composing te relation associated wit te source syncronous strings and te relation associated wit syncronous pair [α, α ], in te most obious way. As in standard constructions tat reduce te rank of a CFG, in tis paper we focus on eac single syncronous rule and factorize it into syncronous rules of lower rank. If we iew te bijectie relation associated wit a syncronous rule as a permutation, we can furter reduce our factorization problem to te problem of factorizing a permutation of arity n into te composition of seeral permutations of arity k < n. Suc factorization can be represented as a tree of composed permutations, called in wat follows a permutation tree. A permutation tree can be conerted into a set of k-ary SCFG rules equialent to te input rule. For example, te input rule: [ X A () B () C () D () E () F () G () H (), X B () A () C () D () G () E () H () F () ] yields te permutation tree of Figure (left). Introducing a new grammar nonterminal X i for eac internal node of te tree yields an equialent set of smaller rules: [ X X () X(), X X() X() ] [ X X () X(), X X () X() ] [ X A () B (), X B () A () ] [ X C () D (), X C () D () ] [ X E () F () G () H (), X G () E () H () F () ] In te case of stocastic grammars, te rule corresponding to te root of te permutation tree is assigned te original rule s probability, wile all oter rules, associated wit new grammar nonterminals, are assigned probability. We process eac rule of an input SCFG independently, producing an equialent grammar wit te smallest possible arity. Factorization Algoritm In tis section we specify and discuss our factorization algoritm. Te algoritm takes as input a permutation defined on te set {,, n}, representing a rule of some SCFG, and proides a permutation tree of arity k n for tat permutation, wit k as small as possible. Permutation trees coering a gien input permutation are unambiguous wit te exception of sequences of binary rules of te same type (eiter inerted or straigt) (Albert et al., 00). Tus, wen factorizing a permutation into a permutation

3 tree, it is safe to greedily reduce a subsequence into a new subtree as soon as a subsequence is found wic represents a continuous span in bot dimensions of te permutation matrix associated wit te input permutation. For space reasons, we omit te proof, but empasize tat any greedy reduction turns out to be eiter necessary, or equialent to te oter alternaties. Any sequences of binary rules can be rearranged into a normalized form (e.g. always leftbrancing) as a postprocessing step, if desired. Te top-leel structure of te algoritm exploits a diide-and-conquer approac, and is te same as tat of te well-known mergesort algoritm (Cormen et al., 990). We work on subsequences of te original permutation, and merge neigboring subsequences into successiely longer subsequences, combining two subsequences of lengt i into a subsequence of lengt i+ until we ae built one subsequence spanning te entire permutation. If eac combination of subsequences can be performed in linear time, ten te entire permutation can be processed in time O(n log n). As in te case of mergesort, tis is an application of te so-called master teorem (Cormen et al., 990). As te algoritm operates, we will maintain te inariant tat we must ae built all subtrees of te target permutation tree tat are entirely witin a gien subsequence tat as been processed. Tis is analogous to te inariant in mergesort tat all processed subsequences are in sorted order. Wen we combine two subsequences, we need only build nodes in te tree tat coer parts of bot subsequences, but are entirely witin te combined subsequence. Tus, we are looking for subtrees tat span te midpoint of te combined subsequence, but ae left and rigt boundaries witin te boundaries of te combined subsequence. In wat follows, tis midpoint is called te split point. From tis inariant, we will be guaranteed to ae a complete, correct permutation tree at te end of last subsequence combination. An example of te operation of te general algoritm is sown in Figure. Te top-leel structure of te algoritm is presented in function KARIZE of Figure. Tere may be more tan one reduction necessary spanning a gien split point wen combining two subsequences. Function MERGE in Fig- A permutation matrix is a way of representing a permutation, and is obtained by rearranging te row (or te columns) of an identity matrix, according to te permutation itself.,,,,,,,,,,,,,,, Figure : Recursie combination of permutation trees. Top row, te input permutation. Second row, after combination into sequences of lengt two, binary nodes ae been built were possible. Tird row, after combination into sequences of lengt four; bottom row, te entire output tree. ure initializes certain data structures described below, and ten cecks for reductions repeatedly until no furter reduction is possible. It looks first for te smallest reduction crossing te split point of te subsequences being combined. If SCAN, described below, finds a alid reduction, it is committed by calling REDUCE. If a reduction is found, we look for furter reductions crossing eiter te left or rigt boundary of te new reduction, repeating until no furter reductions are possible. Because we only need to find reductions spanning te original split point at a gien combination step, tis process is guaranteed to find all reductions needed. We now turn to te problem of identifying a specific reduction to be made across a split point, wic inoles identifying te reduction s left and rigt boundaries. Gien a subsequence and candidate left and rigt boundaries for tat subsequence, te alidity of making a reduction oer tis span can be tested by erifying weter te span constitutes a permuted sequence, tat is, a permutation of a contiguous sequence of integers. Since te starting permutation is defined on a set {,,, n}, we ae no repeated integers in our subsequences, and te aboe condi-

4 function KARIZE() initialize wit identity mapping min max (0.. ); mergesort core for size ; size ; size size * do for min 0; min < -size+; min min + * size do di = min + size - ; max min(, min + *size - ); MERGE(min, di, max); function MERGE(min, di, max) initialize sort [min..max] according to [i]; sort min[min..max] according to [i]; sort max[min..max] according to [i]; merging sorted list takes linear time initialize for i min; i max; i i + do [ [i] ] i; ceck if start of new reduced block if i = min or min[i] min[i-] ten min i; min[ [i] ] min; for i max; i min; i i - do ceck if start of new reduced block if i = max or max[i] max[i+] ten max i ; max[ [i] ] max; look for reductions if SCAN(di) ten REDUCE(scanned reduction); wile SCAN(left) or SCAN(rigt) do REDUCE(smaller reduction); function REDUCE(left, rigt, bot, top) for i bot..top do min[i] left; max[i] rigt; for i left..rigt do min[i] bot; max[i] top; print reduce: left..rigt ; Figure : KARIZE: Top leel of algoritm, identical to tat of mergesort. MERGE: combines two subsequences of size i into new subsequence of size i+. REDUCE: commits reduction by updating min and max arrays. tion can be tested by scanning te span in question, finding te minimum and maximum integers in te span, and cecking weter teir difference is equal to te lengt of te span minus one. Below we call tis condition te reduction test. As an example of te reduction test, consider te subsequence (,,, ), and take te last tree elements, (,, ), as a candidate span. We see tat and are te minimum and maximum integers in te corresponding span, respectiely. We ten compute =, wile te lengt of te span minus one is, implying tat no reduction is possible. Howeer, examining te entire subsequence, te minimum is and te maximum is, and =, wic is te lengt of te span minus one. We terefore conclude tat we can reduce tat span by means of some permutation, tat is, parse te span by means of a node in te permutation tree. Tis reduction constitutes te -ary node in te permutation tree of Figure. A triial implementation of te reduction test would be to tests all combinations of left and rigt boundaries for te new reduction. Unfortunately, tis would take time Ω(n ) for a single subsequence combination step, wereas to aciee te oerall O(n log n) complexity we need linear time for eac combination step. It turns out tat te boundaries of te next reduction, coering a gien split point, can be computed in linear time wit te tecnique sown in function SCAN of Figure. We start wit left and rigt candidate boundaries at te two points immediately to te left and rigt of te split point, and ten repeatedly ceck weter te current left and rigt boundaries identify a permuted sequence by applying te reduction test, and moe te left and rigt boundaries outward as necessary, as soon as missing integers are identified outside te current boundaries, as explained below. We will sow tat, as we moe outward, te number of possible configurations acieed for te positions of te left and te rigt boundaries is linearly bounded in te lengt of te combined subsequence (as opposed to quadratically bounded). In order to efficiently implement te aboe idea, we will in fact maintain four boundaries for te candidate reduction, wic can be isualized as te left, rigt, top and bottom boundaries in te permutation matrix. No explicit representation of te permutation matrix itself is constructed, as tat would require quadratic time. Rater, we

5 Figure : Permutation matrix for input permutation (left) and witin-subsequence permutation (rigt) for subsequences of size four. maintain two arrays:, wic maps from ertical to orizontal positions witin te current subsequence, and wic maps from orizontal to ertical positions. Tese arrays represent te witinsubsequence permutation obtained by sorting te elements of eac subsequence according to te input permutation, wile keeping eac element witin its block, as sown in Figure. Witin eac subsequence, we alternate between scanning orizontally from left to rigt, possibly extending te top and bottom boundaries (Figure lines 9 to ), and scanning ertically from bottom to top, possibly extending te left and rigt boundaries (lines 0 to ). Eac extension is forced wen, looking at te witin-subsequence permutation, we find tat some element is witin te current boundaries in one dimension but outside te boundaries in te oter. If te distance between ertical boundaries is larger in te input permutation tan in te subsequence permutation, necessary elements are missing from te current subsequence and no reduction is possible at tis step (line ). Wen all necessary elements are present in te current subsequence and no furter extensions are necessary to te boundaries (line 0), we ae satisfied te reduction test on te input permutation, and make a reduction. Te trick used to keep te iteratie scanning linear is tat we skip te subsequence scanned on te preious iteration on eac scan, in bot te orizontal and ertical directions. Lines and of Figure perform tis skip by adancing te x and y counters past preiously scanned regions. Considering te orizontal scan of lines 9 to, in a gien iteration of te wile loop, we scan only te items between newleft and left and between rigt and newrigt. On te next iteration of te wile loop, te newleft boundary as moed furter to te left, : function SCAN (di) : left ; : rigt ; : newleft di; : newrigt di + ; : newtop ; : newbot ; : wile do orizontal scan 9: for x newleft; x newrigt ; do 0: newtop max(newtop, max[x]); : newbot min(newbot, min[x]); skip to end of reduced block : x max[min[x]] + ; skip section scanned on last iter : if x = left ten : x rigt + ; : rigt newrigt; : left newleft; te reduction test : if newtop - newbot < : [[newtop]] - [[newbot]] ten 9: return (0); ertical scan 0: for y newbot; y newtop ; do : newrigt : max(newrigt, max[y]); : newleft min(newleft, min[y]); skip to end of reduced block : y max[min[y]] + ; skip section scanned on last iter : if y = bot ten : y top + ; : top newtop; : bot newbot; if no cange to boundaries, reduce 9: if newrigt = rigt 0: and newleft = left ten : return (, left, rigt, bot, top); Figure : Linear time function to ceck for a single reduction at split point di.

6 wile te ariable left takes te preious alue of newleft, ensuring tat te items scanned on tis iteration are distinct from tose already processed. Similarly, on te rigt edge we scan new items, between rigt and newrigt. Te same analysis applies to te ertical scan. Because eac item in te permutation is scanned only once in te ertical direction and once in te orizontal direction, te entire call to SCAN takes linear time, regardless of te number of iterations of te wile loop tat are required. We must furter sow tat eac call to MERGE takes only linear time, despite tat fact tat it may inole many calls to SCAN. We accomplis tis by introducing a second type of skipping in te scans, wic adances past any preiously reduced block in a single step. In order to skip past preious reductions, we maintain (in function REDUCE) auxiliary arrays wit te minimum and maximum positions of te largest block eac point as been reduced to, in bot te orizontal and ertical dimensions. We use tese data structures (min, max, min, max) wen adancing to te next position of te scan in lines and of Figure. Because eac call to SCAN skips items scanned by preious calls, eac item is scanned at most twice across an entire call to MERGE, once wen scanning across a new reduction s left boundary and once wen scanning across te rigt boundary, guaranteeing tat MERGE completes in linear time. An Example In tis section we examine te operation of te algoritm on a permutation of lengt eigt, resulting in te permutation tree of Figure (rigt). We will build up our analysis of te permutation by starting wit indiidual items of te input permutation and building up subsequences of lengt,, and finally. In our example permutation, (,,,,,,, ), no reductions can be made until te final combination step, in wic one permutation of size is used, and one of size. We begin wit te input permutation along te bottom of Figure a. We represent te intermediate data structures, min, and max along te ertical axis of te figure; tese tree arrays are all initialized to be te sequence (,,, ). Figure b sows te combination of indiidual items into subsequences of lengt two. Eac new subsequence of te array is sorted according to a) b) c) max min min max max min min max max min min max Figure : Steps in an example computation, wit input permutation on left and witinsubsequence permutation described by array on rigt. Panel (a) sows initial blocks of unit size, (b) sows combination of unit blocks into blocks of size two, and (c) size two into size four. No reductions are possible in tese stages; example continued in next figure.

7 a) min max min max b) min max min max Left and rigt boundaries are initialized to be adjacent to orizontal split point. Vertical scan sows left and rigt boundaries must be extended. Permutation of size four is reduced. c) min max min max d) min max min max Searc for next reduction: left and rigt boundaries initialized to be adjacent to left edge of preious reduction. Vertical scan sows rigt boundary must be extended. e) min max min max f) min max min max Horizontal scan sows top boundary must be extended. Vertical scan sows left boundary must be extended. Permutation of size fie is reduced. Figure : Steps in scanning for final combination of subsequences, were =. Area witin current left, rigt, top and bottom boundaries is saded; darker sading indicates a reduction. In eac scan, te span scanned in te preious panel is skipped oer.

8 te ertical position of te dots in te corresponding columns. Tus, because [] = > [] =, we swap and in te array. Te algoritm cecks weter any reductions can be made at tis step by computing te difference between te integers on eac side of eac split point. Because none of te pairs of integers in are consecutie, no reductions are made at tis step. Figure c sows te combination te pairs into subsequences of lengt four. Te two split points to be examined are between te second and tird position, and te sixt and seent position. Again, no reductions are possible. Finally we combine te two subsequences of lengt four to complete te analysis of te entire permutation. Te split point is between te fourt and fift positions of te input permutation, and in te first orizontal scan of tese two positions, we see tat [] = and [] =, meaning our top boundary will be and our bottom boundary, sown in Figure a. Scanning ertically from position to, we see orizontal positions,,, and, giing te minimum,, as te new left boundary and te maximum,, as te new rigt boundary, sown in Figure b. We now perform anoter orizontal scan starting at position, but ten jumping directly to position, as orizontal positions and were scanned preiously. After tis scan, te minimum ertical position seen remains, and te maximum ertical position is still. At tis point, because we ae te same boundaries as on te preious scan, we can stop and erify weter te region determined by our current boundaries as te same lengt in te ertical and orizontal dimensions. Bot dimensions ae lengt four, meaning tat we ae found a subsequence tat is continuous in bot dimensions and can safely be reduced, as sown in Figure d. After making tis reduction, we update te min array to ae all s for te newly reduced span, and update max to ae all sixes. We ten ceck weter furter reductions are possible coering tis split point. We repeat te process of scanning orizontally and ertically in Figure c-f, tis time skipping te span just reduced. One furter reduction is possible, coering te entire input permutation, as sown in Figure f. Conclusion Te algoritm aboe not only identifies weter a permutation can be factored into a composition of permutations, but also returns te factorization tat minimizes te largest rule size, in time O(n log n). Te factored SCFG wit rules of size at most k can be used to syncronously parse in time O(N k+ ) by dynamic programming wit continuous spans in one dimension. As mentioned in te introduction, te optimal parsing strategy for SCFG rules wit a gien permutation may inole dynamic programming states wit discontinuous spans in bot dimensions. Weter tese optimal parsing strategies can be found efficiently remains an interesting open problem. Acknowledgments Tis work was partially supported by NSF ITR IIS-09 and NSF ITR IIS-000. References Albert V. Ao and Jeffery D. Ullman. 9. Te Teory of Parsing, Translation, and Compiling, olume. Prentice-Hall, Englewood Cliffs, NJ. M. H. Albert, M. D. Atkinson, and M. Klazar. 00. Te enumeration of simple permutations. Journal of Integer Sequences, (0..): pages. Daid Ciang. 00. A ierarcical prase-based model for statistical macine translation. In Proceedings of ACL-0, pages 0. Tomas H. Cormen, Carles E. Leiserson, and Ronald L. Riest Introduction to algoritms. MIT Press, Cambridge, MA. Micel Galley, Mark Hopkins, Kein Knigt, and Daniel Marcu. 00. Wat s in a translation rule? In Proceedings of HLT/NAACL. I. Dan Melamed. 00. Multitext grammars and syncronous parsers. In Proceedings of HLT/NAACL. Giorgio Satta and Enoc Peserico. 00. Some computational complexity results for syncronous context-free grammars. In Proceedings of HLT/EMNLP, pages 0 0. Dekai Wu. 99. Stocastic inersion transduction grammars and bilingual parsing of parallel corpora. Computational Linguistics, (): 0. Kenji Yamada and Kein Knigt. 00. A syntaxbased statistical translation model. In Proceedings of ACL-0. Hao Zang, Liang Huang, Daniel Gildea, and Kein Knigt. 00. Syncronous binarization for macine translation. In Proceedings of HLT/NAACL.

Research on the Anti-perspective Correction Algorithm of QR Barcode

Research on the Anti-perspective Correction Algorithm of QR Barcode Researc on te Anti-perspective Correction Algoritm of QR Barcode Jianua Li, Yi-Wen Wang, YiJun Wang,Yi Cen, Guoceng Wang Key Laboratory of Electronic Tin Films and Integrated Devices University of Electronic

More information

M(0) = 1 M(1) = 2 M(h) = M(h 1) + M(h 2) + 1 (h > 1)

M(0) = 1 M(1) = 2 M(h) = M(h 1) + M(h 2) + 1 (h > 1) Insertion and Deletion in VL Trees Submitted in Partial Fulfillment of te Requirements for Dr. Eric Kaltofen s 66621: nalysis of lgoritms by Robert McCloskey December 14, 1984 1 ackground ccording to Knut

More information

The EOQ Inventory Formula

The EOQ Inventory Formula Te EOQ Inventory Formula James M. Cargal Matematics Department Troy University Montgomery Campus A basic problem for businesses and manufacturers is, wen ordering supplies, to determine wat quantity of

More information

Verifying Numerical Convergence Rates

Verifying Numerical Convergence Rates 1 Order of accuracy Verifying Numerical Convergence Rates We consider a numerical approximation of an exact value u. Te approximation depends on a small parameter, suc as te grid size or time step, and

More information

An inquiry into the multiplier process in IS-LM model

An inquiry into the multiplier process in IS-LM model An inquiry into te multiplier process in IS-LM model Autor: Li ziran Address: Li ziran, Room 409, Building 38#, Peing University, Beijing 00.87,PRC. Pone: (86) 00-62763074 Internet Address: jefferson@water.pu.edu.cn

More information

Synchronous Binarization for Machine Translation

Synchronous Binarization for Machine Translation Synchronous Binarization for Machine Translation Hao Zhang Computer Science Department University of Rochester Rochester, NY 14627 zhanghao@cs.rochester.edu Daniel Gildea Computer Science Department University

More information

How To Ensure That An Eac Edge Program Is Successful

How To Ensure That An Eac Edge Program Is Successful Introduction Te Economic Diversification and Growt Enterprises Act became effective on 1 January 1995. Te creation of tis Act was to encourage new businesses to start or expand in Newfoundland and Labrador.

More information

Optimized Data Indexing Algorithms for OLAP Systems

Optimized Data Indexing Algorithms for OLAP Systems Database Systems Journal vol. I, no. 2/200 7 Optimized Data Indexing Algoritms for OLAP Systems Lucian BORNAZ Faculty of Cybernetics, Statistics and Economic Informatics Academy of Economic Studies, Bucarest

More information

Tangent Lines and Rates of Change

Tangent Lines and Rates of Change Tangent Lines and Rates of Cange 9-2-2005 Given a function y = f(x), ow do you find te slope of te tangent line to te grap at te point P(a, f(a))? (I m tinking of te tangent line as a line tat just skims

More information

SAT Subject Math Level 1 Facts & Formulas

SAT Subject Math Level 1 Facts & Formulas Numbers, Sequences, Factors Integers:..., -3, -2, -1, 0, 1, 2, 3,... Reals: integers plus fractions, decimals, and irrationals ( 2, 3, π, etc.) Order Of Operations: Aritmetic Sequences: PEMDAS (Parenteses

More information

- 1 - Handout #22 May 23, 2012 Huffman Encoding and Data Compression. CS106B Spring 2012. Handout by Julie Zelenski with minor edits by Keith Schwarz

- 1 - Handout #22 May 23, 2012 Huffman Encoding and Data Compression. CS106B Spring 2012. Handout by Julie Zelenski with minor edits by Keith Schwarz CS106B Spring 01 Handout # May 3, 01 Huffman Encoding and Data Compression Handout by Julie Zelenski wit minor edits by Keit Scwarz In te early 1980s, personal computers ad ard disks tat were no larger

More information

Schedulability Analysis under Graph Routing in WirelessHART Networks

Schedulability Analysis under Graph Routing in WirelessHART Networks Scedulability Analysis under Grap Routing in WirelessHART Networks Abusayeed Saifulla, Dolvara Gunatilaka, Paras Tiwari, Mo Sa, Cenyang Lu, Bo Li Cengjie Wu, and Yixin Cen Department of Computer Science,

More information

ACT Math Facts & Formulas

ACT Math Facts & Formulas Numbers, Sequences, Factors Integers:..., -3, -2, -1, 0, 1, 2, 3,... Rationals: fractions, tat is, anyting expressable as a ratio of integers Reals: integers plus rationals plus special numbers suc as

More information

Improved dynamic programs for some batcing problems involving te maximum lateness criterion A P M Wagelmans Econometric Institute Erasmus University Rotterdam PO Box 1738, 3000 DR Rotterdam Te Neterlands

More information

FINITE DIFFERENCE METHODS

FINITE DIFFERENCE METHODS FINITE DIFFERENCE METHODS LONG CHEN Te best known metods, finite difference, consists of replacing eac derivative by a difference quotient in te classic formulation. It is simple to code and economic to

More information

Recall from last time: Events are recorded by local observers with synchronized clocks. Event 1 (firecracker explodes) occurs at x=x =0 and t=t =0

Recall from last time: Events are recorded by local observers with synchronized clocks. Event 1 (firecracker explodes) occurs at x=x =0 and t=t =0 1/27 Day 5: Questions? Time Dilation engt Contraction PH3 Modern Pysics P11 I sometimes ask myself ow it came about tat I was te one to deelop te teory of relatiity. Te reason, I tink, is tat a normal

More information

Computer Science and Engineering, UCSD October 7, 1999 Goldreic-Levin Teorem Autor: Bellare Te Goldreic-Levin Teorem 1 Te problem We æx a an integer n for te lengt of te strings involved. If a is an n-bit

More information

Distances in random graphs with infinite mean degrees

Distances in random graphs with infinite mean degrees Distances in random graps wit infinite mean degrees Henri van den Esker, Remco van der Hofstad, Gerard Hoogiemstra and Dmitri Znamenski April 26, 2005 Abstract We study random graps wit an i.i.d. degree

More information

Comparison between two approaches to overload control in a Real Server: local or hybrid solutions?

Comparison between two approaches to overload control in a Real Server: local or hybrid solutions? Comparison between two approaces to overload control in a Real Server: local or ybrid solutions? S. Montagna and M. Pignolo Researc and Development Italtel S.p.A. Settimo Milanese, ITALY Abstract Tis wor

More information

2 Limits and Derivatives

2 Limits and Derivatives 2 Limits and Derivatives 2.7 Tangent Lines, Velocity, and Derivatives A tangent line to a circle is a line tat intersects te circle at exactly one point. We would like to take tis idea of tangent line

More information

Can a Lump-Sum Transfer Make Everyone Enjoy the Gains. from Free Trade?

Can a Lump-Sum Transfer Make Everyone Enjoy the Gains. from Free Trade? Can a Lump-Sum Transfer Make Everyone Enjoy te Gains from Free Trade? Yasukazu Icino Department of Economics, Konan University June 30, 2010 Abstract I examine lump-sum transfer rules to redistribute te

More information

A system to monitor the quality of automated coding of textual answers to open questions

A system to monitor the quality of automated coding of textual answers to open questions Researc in Official Statistics Number 2/2001 A system to monitor te quality of automated coding of textual answers to open questions Stefania Maccia * and Marcello D Orazio ** Italian National Statistical

More information

SWITCH T F T F SELECT. (b) local schedule of two branches. (a) if-then-else construct A & B MUX. one iteration cycle

SWITCH T F T F SELECT. (b) local schedule of two branches. (a) if-then-else construct A & B MUX. one iteration cycle 768 IEEE RANSACIONS ON COMPUERS, VOL. 46, NO. 7, JULY 997 Compile-ime Sceduling of Dynamic Constructs in Dataæow Program Graps Soonoi Ha, Member, IEEE and Edward A. Lee, Fellow, IEEE Abstract Sceduling

More information

Math 113 HW #5 Solutions

Math 113 HW #5 Solutions Mat 3 HW #5 Solutions. Exercise.5.6. Suppose f is continuous on [, 5] and te only solutions of te equation f(x) = 6 are x = and x =. If f() = 8, explain wy f(3) > 6. Answer: Suppose we ad tat f(3) 6. Ten

More information

Geometric Stratification of Accounting Data

Geometric Stratification of Accounting Data Stratification of Accounting Data Patricia Gunning * Jane Mary Horgan ** William Yancey *** Abstract: We suggest a new procedure for defining te boundaries of te strata in igly skewed populations, usual

More information

f(a + h) f(a) f (a) = lim

f(a + h) f(a) f (a) = lim Lecture 7 : Derivative AS a Function In te previous section we defined te derivative of a function f at a number a (wen te function f is defined in an open interval containing a) to be f (a) 0 f(a + )

More information

OPTIMAL DISCONTINUOUS GALERKIN METHODS FOR THE ACOUSTIC WAVE EQUATION IN HIGHER DIMENSIONS

OPTIMAL DISCONTINUOUS GALERKIN METHODS FOR THE ACOUSTIC WAVE EQUATION IN HIGHER DIMENSIONS OPTIMAL DISCONTINUOUS GALERKIN METHODS FOR THE ACOUSTIC WAVE EQUATION IN HIGHER DIMENSIONS ERIC T. CHUNG AND BJÖRN ENGQUIST Abstract. In tis paper, we developed and analyzed a new class of discontinuous

More information

Derivatives Math 120 Calculus I D Joyce, Fall 2013

Derivatives Math 120 Calculus I D Joyce, Fall 2013 Derivatives Mat 20 Calculus I D Joyce, Fall 203 Since we ave a good understanding of its, we can develop derivatives very quickly. Recall tat we defined te derivative f x of a function f at x to be te

More information

Instantaneous Rate of Change:

Instantaneous Rate of Change: Instantaneous Rate of Cange: Last section we discovered tat te average rate of cange in F(x) can also be interpreted as te slope of a scant line. Te average rate of cange involves te cange in F(x) over

More information

Strategic trading in a dynamic noisy market. Dimitri Vayanos

Strategic trading in a dynamic noisy market. Dimitri Vayanos LSE Researc Online Article (refereed) Strategic trading in a dynamic noisy market Dimitri Vayanos LSE as developed LSE Researc Online so tat users may access researc output of te Scool. Copyrigt and Moral

More information

Math Test Sections. The College Board: Expanding College Opportunity

Math Test Sections. The College Board: Expanding College Opportunity Taking te SAT I: Reasoning Test Mat Test Sections Te materials in tese files are intended for individual use by students getting ready to take an SAT Program test; permission for any oter use must be sougt

More information

Note nine: Linear programming CSE 101. 1 Linear constraints and objective functions. 1.1 Introductory example. Copyright c Sanjoy Dasgupta 1

Note nine: Linear programming CSE 101. 1 Linear constraints and objective functions. 1.1 Introductory example. Copyright c Sanjoy Dasgupta 1 Copyrigt c Sanjoy Dasgupta Figure. (a) Te feasible region for a linear program wit two variables (see tet for details). (b) Contour lines of te objective function: for different values of (profit). Te

More information

schema binary search tree schema binary search trees data structures and algorithms 2015 09 21 lecture 7 AVL-trees material

schema binary search tree schema binary search trees data structures and algorithms 2015 09 21 lecture 7 AVL-trees material scema binary searc trees data structures and algoritms 05 0 lecture 7 VL-trees material scema binary searc tree binary tree: linked data structure wit nodes containing binary searc trees VL-trees material

More information

OPTIMAL FLEET SELECTION FOR EARTHMOVING OPERATIONS

OPTIMAL FLEET SELECTION FOR EARTHMOVING OPERATIONS New Developments in Structural Engineering and Construction Yazdani, S. and Sing, A. (eds.) ISEC-7, Honolulu, June 18-23, 2013 OPTIMAL FLEET SELECTION FOR EARTHMOVING OPERATIONS JIALI FU 1, ERIK JENELIUS

More information

SAMPLE DESIGN FOR THE TERRORISM RISK INSURANCE PROGRAM SURVEY

SAMPLE DESIGN FOR THE TERRORISM RISK INSURANCE PROGRAM SURVEY ASA Section on Survey Researc Metods SAMPLE DESIG FOR TE TERRORISM RISK ISURACE PROGRAM SURVEY G. ussain Coudry, Westat; Mats yfjäll, Statisticon; and Marianne Winglee, Westat G. ussain Coudry, Westat,

More information

Binary Search Trees. Adnan Aziz. Heaps can perform extract-max, insert efficiently O(log n) worst case

Binary Search Trees. Adnan Aziz. Heaps can perform extract-max, insert efficiently O(log n) worst case Binary Searc Trees Adnan Aziz 1 BST basics Based on CLRS, C 12. Motivation: Heaps can perform extract-max, insert efficiently O(log n) worst case Has tables can perform insert, delete, lookup efficiently

More information

Lecture 10: What is a Function, definition, piecewise defined functions, difference quotient, domain of a function

Lecture 10: What is a Function, definition, piecewise defined functions, difference quotient, domain of a function Lecture 10: Wat is a Function, definition, piecewise defined functions, difference quotient, domain of a function A function arises wen one quantity depends on anoter. Many everyday relationsips between

More information

SAT Math Must-Know Facts & Formulas

SAT Math Must-Know Facts & Formulas SAT Mat Must-Know Facts & Formuas Numbers, Sequences, Factors Integers:..., -3, -2, -1, 0, 1, 2, 3,... Rationas: fractions, tat is, anyting expressabe as a ratio of integers Reas: integers pus rationas

More information

Unemployment insurance/severance payments and informality in developing countries

Unemployment insurance/severance payments and informality in developing countries Unemployment insurance/severance payments and informality in developing countries David Bardey y and Fernando Jaramillo z First version: September 2011. Tis version: November 2011. Abstract We analyze

More information

2.23 Gambling Rehabilitation Services. Introduction

2.23 Gambling Rehabilitation Services. Introduction 2.23 Gambling Reabilitation Services Introduction Figure 1 Since 1995 provincial revenues from gambling activities ave increased over 56% from $69.2 million in 1995 to $108 million in 2004. Te majority

More information

Empirical Evaluation of Overlap Requirements of Adjacent Radio Cells for Zero Delay Handover

Empirical Evaluation of Overlap Requirements of Adjacent Radio Cells for Zero Delay Handover Empirical Ealuation of Oerlap Requirements of Adjacent Radio Cells for Zero Delay Handoer Marc Emmelmann and Hyung-Taek Lim Tecnical Uniersity Berlin Telecommunication Networks Group TKN Berlin, Germany

More information

Note: Principal version Modification Modification Complete version from 1 October 2014 Business Law Corporate and Contract Law

Note: Principal version Modification Modification Complete version from 1 October 2014 Business Law Corporate and Contract Law Note: Te following curriculum is a consolidated version. It is legally non-binding and for informational purposes only. Te legally binding versions are found in te University of Innsbruck Bulletins (in

More information

LZ77. Example 2.10: Let T = badadadabaab and assume d max and l max are large. phrase b a d adadab aa b

LZ77. Example 2.10: Let T = badadadabaab and assume d max and l max are large. phrase b a d adadab aa b LZ77 The original LZ77 algorithm works as follows: A phrase T j starting at a position i is encoded as a triple of the form distance, length, symbol. A triple d, l, s means that: T j = T [i...i + l] =

More information

Training Robust Support Vector Regression via D. C. Program

Training Robust Support Vector Regression via D. C. Program Journal of Information & Computational Science 7: 12 (2010) 2385 2394 Available at ttp://www.joics.com Training Robust Support Vector Regression via D. C. Program Kuaini Wang, Ping Zong, Yaoong Zao College

More information

New Vocabulary volume

New Vocabulary volume -. Plan Objectives To find te volume of a prism To find te volume of a cylinder Examples Finding Volume of a Rectangular Prism Finding Volume of a Triangular Prism 3 Finding Volume of a Cylinder Finding

More information

Binary Heaps * * * * * * * / / \ / \ / \ / \ / \ * * * * * * * * * * * / / \ / \ / / \ / \ * * * * * * * * * *

Binary Heaps * * * * * * * / / \ / \ / \ / \ / \ * * * * * * * * * * * / / \ / \ / / \ / \ * * * * * * * * * * Binary Heaps A binary heap is another data structure. It implements a priority queue. Priority Queue has the following operations: isempty add (with priority) remove (highest priority) peek (at highest

More information

Notes: Most of the material in this chapter is taken from Young and Freedman, Chap. 12.

Notes: Most of the material in this chapter is taken from Young and Freedman, Chap. 12. Capter 6. Fluid Mecanics Notes: Most of te material in tis capter is taken from Young and Freedman, Cap. 12. 6.1 Fluid Statics Fluids, i.e., substances tat can flow, are te subjects of tis capter. But

More information

To motivate the notion of a variogram for a covariance stationary process, { Ys ( ): s R}

To motivate the notion of a variogram for a covariance stationary process, { Ys ( ): s R} 4. Variograms Te covariogram and its normalized form, te correlogram, are by far te most intuitive metods for summarizing te structure of spatial dependencies in a covariance stationary process. However,

More information

Chapter 11. Limits and an Introduction to Calculus. Selected Applications

Chapter 11. Limits and an Introduction to Calculus. Selected Applications Capter Limits and an Introduction to Calculus. Introduction to Limits. Tecniques for Evaluating Limits. Te Tangent Line Problem. Limits at Infinit and Limits of Sequences.5 Te Area Problem Selected Applications

More information

IBMEaseofUse. User Interface Architecture

IBMEaseofUse. User Interface Architecture IBMEaseofse ser Interface Arcitecture IBMEaseofse ser Interface Arcitecture Note Before using tis information and te product it supports, read te information in Notices on page 33. Second Edition (December,

More information

Cyber Epidemic Models with Dependences

Cyber Epidemic Models with Dependences Cyber Epidemic Models wit Dependences Maocao Xu 1, Gaofeng Da 2 and Souuai Xu 3 1 Department of Matematics, Illinois State University mxu2@ilstu.edu 2 Institute for Cyber Security, University of Texas

More information

1.6. Analyse Optimum Volume and Surface Area. Maximum Volume for a Given Surface Area. Example 1. Solution

1.6. Analyse Optimum Volume and Surface Area. Maximum Volume for a Given Surface Area. Example 1. Solution 1.6 Analyse Optimum Volume and Surface Area Estimation and oter informal metods of optimizing measures suc as surface area and volume often lead to reasonable solutions suc as te design of te tent in tis

More information

Area-Specific Recreation Use Estimation Using the National Visitor Use Monitoring Program Data

Area-Specific Recreation Use Estimation Using the National Visitor Use Monitoring Program Data United States Department of Agriculture Forest Service Pacific Nortwest Researc Station Researc Note PNW-RN-557 July 2007 Area-Specific Recreation Use Estimation Using te National Visitor Use Monitoring

More information

The modelling of business rules for dashboard reporting using mutual information

The modelling of business rules for dashboard reporting using mutual information 8 t World IMACS / MODSIM Congress, Cairns, Australia 3-7 July 2009 ttp://mssanz.org.au/modsim09 Te modelling of business rules for dasboard reporting using mutual information Gregory Calbert Command, Control,

More information

Average and Instantaneous Rates of Change: The Derivative

Average and Instantaneous Rates of Change: The Derivative 9.3 verage and Instantaneous Rates of Cange: Te Derivative 609 OBJECTIVES 9.3 To define and find average rates of cange To define te derivative as a rate of cange To use te definition of derivative to

More information

Determine the perimeter of a triangle using algebra Find the area of a triangle using the formula

Determine the perimeter of a triangle using algebra Find the area of a triangle using the formula Student Name: Date: Contact Person Name: Pone Number: Lesson 0 Perimeter, Area, and Similarity of Triangles Objectives Determine te perimeter of a triangle using algebra Find te area of a triangle using

More information

2.12 Student Transportation. Introduction

2.12 Student Transportation. Introduction Introduction Figure 1 At 31 Marc 2003, tere were approximately 84,000 students enrolled in scools in te Province of Newfoundland and Labrador, of wic an estimated 57,000 were transported by scool buses.

More information

Chapter 7 Numerical Differentiation and Integration

Chapter 7 Numerical Differentiation and Integration 45 We ave a abit in writing articles publised in scientiþc journals to make te work as Þnised as possible, to cover up all te tracks, to not worry about te blind alleys or describe ow you ad te wrong idea

More information

In other words the graph of the polynomial should pass through the points

In other words the graph of the polynomial should pass through the points Capter 3 Interpolation Interpolation is te problem of fitting a smoot curve troug a given set of points, generally as te grap of a function. It is useful at least in data analysis (interpolation is a form

More information

Staffing and routing in a two-tier call centre. Sameer Hasija*, Edieal J. Pinker and Robert A. Shumsky

Staffing and routing in a two-tier call centre. Sameer Hasija*, Edieal J. Pinker and Robert A. Shumsky 8 Int. J. Operational Researc, Vol. 1, Nos. 1/, 005 Staffing and routing in a two-tier call centre Sameer Hasija*, Edieal J. Pinker and Robert A. Sumsky Simon Scool, University of Rocester, Rocester 1467,

More information

The Derivative as a Function

The Derivative as a Function Section 2.2 Te Derivative as a Function 200 Kiryl Tsiscanka Te Derivative as a Function DEFINITION: Te derivative of a function f at a number a, denoted by f (a), is if tis limit exists. f (a) f(a+) f(a)

More information

Catalogue no. 12-001-XIE. Survey Methodology. December 2004

Catalogue no. 12-001-XIE. Survey Methodology. December 2004 Catalogue no. 1-001-XIE Survey Metodology December 004 How to obtain more information Specific inquiries about tis product and related statistics or services sould be directed to: Business Survey Metods

More information

SAT Math Facts & Formulas

SAT Math Facts & Formulas Numbers, Sequences, Factors SAT Mat Facts & Formuas Integers:..., -3, -2, -1, 0, 1, 2, 3,... Reas: integers pus fractions, decimas, and irrationas ( 2, 3, π, etc.) Order Of Operations: Aritmetic Sequences:

More information

Section 2.3 Solving Right Triangle Trigonometry

Section 2.3 Solving Right Triangle Trigonometry Section.3 Solving Rigt Triangle Trigonometry Eample In te rigt triangle ABC, A = 40 and c = 1 cm. Find a, b, and B. sin 40 a a c 1 a 1sin 40 7.7cm cos 40 b c b 1 b 1cos40 9.cm A 40 1 b C B a B = 90 - A

More information

Pre-trial Settlement with Imperfect Private Monitoring

Pre-trial Settlement with Imperfect Private Monitoring Pre-trial Settlement wit Imperfect Private Monitoring Mostafa Beskar University of New Hampsire Jee-Hyeong Park y Seoul National University July 2011 Incomplete, Do Not Circulate Abstract We model pretrial

More information

Once you have reviewed the bulletin, please

Once you have reviewed the bulletin, please Akron Public Scools OFFICE OF CAREER EDUCATION BULLETIN #5 : Driver Responsibilities 1. Akron Board of Education employees assigned to drive Board-owned or leased veicles MUST BE FAMILIAR wit te Business

More information

Writing Mathematics Papers

Writing Mathematics Papers Writing Matematics Papers Tis essay is intended to elp your senior conference paper. It is a somewat astily produced amalgam of advice I ave given to students in my PDCs (Mat 4 and Mat 9), so it s not

More information

A strong credit score can help you score a lower rate on a mortgage

A strong credit score can help you score a lower rate on a mortgage NET GAIN Scoring points for your financial future AS SEEN IN USA TODAY S MONEY SECTION, JULY 3, 2007 A strong credit score can elp you score a lower rate on a mortgage By Sandra Block Sales of existing

More information

Torchmark Corporation 2001 Third Avenue South Birmingham, Alabama 35233 Contact: Joyce Lane 972-569-3627 NYSE Symbol: TMK

Torchmark Corporation 2001 Third Avenue South Birmingham, Alabama 35233 Contact: Joyce Lane 972-569-3627 NYSE Symbol: TMK News Release Torcmark Corporation 2001 Tird Avenue Sout Birmingam, Alabama 35233 Contact: Joyce Lane 972-569-3627 NYSE Symbol: TMK TORCHMARK CORPORATION REPORTS FOURTH QUARTER AND YEAR-END 2004 RESULTS

More information

CHAPTER 7. Di erentiation

CHAPTER 7. Di erentiation CHAPTER 7 Di erentiation 1. Te Derivative at a Point Definition 7.1. Let f be a function defined on a neigborood of x 0. f is di erentiable at x 0, if te following it exists: f 0 fx 0 + ) fx 0 ) x 0 )=.

More information

Shell and Tube Heat Exchanger

Shell and Tube Heat Exchanger Sell and Tube Heat Excanger MECH595 Introduction to Heat Transfer Professor M. Zenouzi Prepared by: Andrew Demedeiros, Ryan Ferguson, Bradford Powers November 19, 2009 1 Abstract 2 Contents Discussion

More information

3 Ans. 1 of my $30. 3 on. 1 on ice cream and the rest on 2011 MATHCOUNTS STATE COMPETITION SPRINT ROUND

3 Ans. 1 of my $30. 3 on. 1 on ice cream and the rest on 2011 MATHCOUNTS STATE COMPETITION SPRINT ROUND 0 MATHCOUNTS STATE COMPETITION SPRINT ROUND. boy scouts are accompanied by scout leaders. Eac person needs bottles of water per day and te trip is day. + = 5 people 5 = 5 bottles Ans.. Cammie as pennies,

More information

For Sale By Owner Program. We can help with our for sale by owner kit that includes:

For Sale By Owner Program. We can help with our for sale by owner kit that includes: Dawn Coen Broker/Owner For Sale By Owner Program If you want to sell your ome By Owner wy not:: For Sale Dawn Coen Broker/Owner YOUR NAME YOUR PHONE # Look as professional as possible Be totally prepared

More information

Multigrid computational methods are

Multigrid computational methods are M ULTIGRID C OMPUTING Wy Multigrid Metods Are So Efficient Originally introduced as a way to numerically solve elliptic boundary-value problems, multigrid metods, and teir various multiscale descendants,

More information

College Planning Using Cash Value Life Insurance

College Planning Using Cash Value Life Insurance College Planning Using Cas Value Life Insurance CAUTION: Te advisor is urged to be extremely cautious of anoter college funding veicle wic provides a guaranteed return of premium immediately if funded

More information

Neighborhood-Privacy Protected Shortest Distance Computing in Cloud

Neighborhood-Privacy Protected Shortest Distance Computing in Cloud Neigboroo-Priacy Protecte Sortest Distance Computing in Clou Jun Gao Jeery Yu Xu Ruoming Jin Jiasuai Zou Tengjiao Wang Dongqing Yang Key Laboratory o Hig Conience Sotware Tecnologies, EECS, Peking Uniersity

More information

What is Advanced Corporate Finance? What is finance? What is Corporate Finance? Deciding how to optimally manage a firm s assets and liabilities.

What is Advanced Corporate Finance? What is finance? What is Corporate Finance? Deciding how to optimally manage a firm s assets and liabilities. Wat is? Spring 2008 Note: Slides are on te web Wat is finance? Deciding ow to optimally manage a firm s assets and liabilities. Managing te costs and benefits associated wit te timing of cas in- and outflows

More information

Projective Geometry. Projective Geometry

Projective Geometry. Projective Geometry Euclidean versus Euclidean geometry describes sapes as tey are Properties of objects tat are uncanged by rigid motions» Lengts» Angles» Parallelism Projective geometry describes objects as tey appear Lengts,

More information

Tis Problem and Retail Inventory Management

Tis Problem and Retail Inventory Management Optimizing Inventory Replenisment of Retail Fasion Products Marsall Fiser Kumar Rajaram Anant Raman Te Warton Scool, University of Pennsylvania, 3620 Locust Walk, 3207 SH-DH, Piladelpia, Pennsylvania 19104-6366

More information

Pretrial Settlement with Imperfect Private Monitoring

Pretrial Settlement with Imperfect Private Monitoring Pretrial Settlement wit Imperfect Private Monitoring Mostafa Beskar Indiana University Jee-Hyeong Park y Seoul National University April, 2016 Extremely Preliminary; Please Do Not Circulate. Abstract We

More information

Predicting the behavior of interacting humans by fusing data from multiple sources

Predicting the behavior of interacting humans by fusing data from multiple sources Predicting te beavior of interacting umans by fusing data from multiple sources Erik J. Sclict 1, Ritcie Lee 2, David H. Wolpert 3,4, Mykel J. Kocenderfer 1, and Brendan Tracey 5 1 Lincoln Laboratory,

More information

Channel Allocation in Non-Cooperative Multi-Radio Multi-Channel Wireless Networks

Channel Allocation in Non-Cooperative Multi-Radio Multi-Channel Wireless Networks Cannel Allocation in Non-Cooperative Multi-Radio Multi-Cannel Wireless Networks Dejun Yang, Xi Fang, Guoliang Xue Arizona State University Abstract Wile tremendous efforts ave been made on cannel allocation

More information

Analyzing the Effects of Insuring Health Risks:

Analyzing the Effects of Insuring Health Risks: Analyzing te Effects of Insuring Healt Risks: On te Trade-off between Sort Run Insurance Benefits vs. Long Run Incentive Costs Harold L. Cole University of Pennsylvania and NBER Soojin Kim University of

More information

A mathematical analysis of the influence of wind uncertainty on MTCD efficiency

A mathematical analysis of the influence of wind uncertainty on MTCD efficiency A mathematical analysis of the influence of wind uncertainty on MTC efficiency Jean-Marc Alliot (SNA/R&) Nicolas urand (SNA/R&) Abstract A large part of the controller s workload comes from conflict detection

More information

ON LOCAL LIKELIHOOD DENSITY ESTIMATION WHEN THE BANDWIDTH IS LARGE

ON LOCAL LIKELIHOOD DENSITY ESTIMATION WHEN THE BANDWIDTH IS LARGE ON LOCAL LIKELIHOOD DENSITY ESTIMATION WHEN THE BANDWIDTH IS LARGE Byeong U. Park 1 and Young Kyung Lee 2 Department of Statistics, Seoul National University, Seoul, Korea Tae Yoon Kim 3 and Ceolyong Park

More information

CHAPTER TWO. f(x) Slope = f (3) = Rate of change of f at 3. x 3. f(1.001) f(1) Average velocity = 1.1 1 1.01 1. s(0.8) s(0) 0.8 0

CHAPTER TWO. f(x) Slope = f (3) = Rate of change of f at 3. x 3. f(1.001) f(1) Average velocity = 1.1 1 1.01 1. s(0.8) s(0) 0.8 0 CHAPTER TWO 2.1 SOLUTIONS 99 Solutions for Section 2.1 1. (a) Te average rate of cange is te slope of te secant line in Figure 2.1, wic sows tat tis slope is positive. (b) Te instantaneous rate of cange

More information

Pressure. Pressure. Atmospheric pressure. Conceptual example 1: Blood pressure. Pressure is force per unit area:

Pressure. Pressure. Atmospheric pressure. Conceptual example 1: Blood pressure. Pressure is force per unit area: Pressure Pressure is force per unit area: F P = A Pressure Te direction of te force exerted on an object by a fluid is toward te object and perpendicular to its surface. At a microscopic level, te force

More information

Simultaneous Location of Trauma Centers and Helicopters for Emergency Medical Service Planning

Simultaneous Location of Trauma Centers and Helicopters for Emergency Medical Service Planning Simultaneous Location of Trauma Centers and Helicopters for Emergency Medical Service Planning Soo-Haeng Co Hoon Jang Taesik Lee Jon Turner Tepper Scool of Business, Carnegie Mellon University, Pittsburg,

More information

On a Satellite Coverage

On a Satellite Coverage I. INTRODUCTION On a Satellite Coverage Problem DANNY T. CHI Kodak Berkeley Researc Yu T. su National Ciao Tbng University Te eart coverage area for a satellite in an Eart syncronous orbit wit a nonzero

More information

1 The Collocation Method

1 The Collocation Method CS410 Assignment 7 Due: 1/5/14 (Fri) at 6pm You must wor eiter on your own or wit one partner. You may discuss bacground issues and general solution strategies wit oters, but te solutions you submit must

More information

Referendum-led Immigration Policy in the Welfare State

Referendum-led Immigration Policy in the Welfare State Referendum-led Immigration Policy in te Welfare State YUJI TAMURA Department of Economics, University of Warwick, UK First version: 12 December 2003 Updated: 16 Marc 2004 Abstract Preferences of eterogeneous

More information

MULTY BINARY TURBO CODED WOFDM PERFORMANCE IN FLAT RAYLEIGH FADING CHANNELS

MULTY BINARY TURBO CODED WOFDM PERFORMANCE IN FLAT RAYLEIGH FADING CHANNELS Volume 49, Number 3, 28 MULTY BINARY TURBO CODED WOFDM PERFORMANCE IN FLAT RAYLEIGH FADING CHANNELS Marius OLTEAN Maria KOVACI Horia BALTA Andrei CAMPEANU Faculty of, Timisoara, Romania Bd. V. Parvan,

More information

KM client format supported by KB valid from 13 May 2015

KM client format supported by KB valid from 13 May 2015 supported by KB valid from 13 May 2015 1/16 VRSION 1.2. UPDATD: 13.12.2005. 1 Introduction... 2 1.1 Purpose of tis document... 2 1.2 Caracteristics of te KM format... 2 2 Formal ceck of KM format... 3

More information

Design and Analysis of a Fault-Tolerant Mechanism for a Server-Less Video-On-Demand System

Design and Analysis of a Fault-Tolerant Mechanism for a Server-Less Video-On-Demand System Design and Analysis of a Fault-olerant Mecanism for a Server-Less Video-On-Demand System Jack Y. B. Lee Department of Information Engineering e Cinese University of Hong Kong Satin, N.., Hong Kong Email:

More information

Strategic trading and welfare in a dynamic market. Dimitri Vayanos

Strategic trading and welfare in a dynamic market. Dimitri Vayanos LSE Researc Online Article (refereed) Strategic trading and welfare in a dynamic market Dimitri Vayanos LSE as developed LSE Researc Online so tat users may access researc output of te Scool. Copyrigt

More information

2.1: The Derivative and the Tangent Line Problem

2.1: The Derivative and the Tangent Line Problem .1.1.1: Te Derivative and te Tangent Line Problem Wat is te deinition o a tangent line to a curve? To answer te diiculty in writing a clear deinition o a tangent line, we can deine it as te iting position

More information

Scrambling in German { Extraction into te Mittelfeld Stefan Muller y Humboldt Universitat zu Berlin August, 199 Abstract German is a language wit a relatively free word order. During te last few years

More information

Sections 3.1/3.2: Introducing the Derivative/Rules of Differentiation

Sections 3.1/3.2: Introducing the Derivative/Rules of Differentiation Sections 3.1/3.2: Introucing te Derivative/Rules of Differentiation 1 Tangent Line Before looking at te erivative, refer back to Section 2.1, looking at average velocity an instantaneous velocity. Here

More information

Probabilistic Community and Role Model for Social Networks

Probabilistic Community and Role Model for Social Networks Probabilistic Community and Role Model for Social Networks Yu Han and Jie Tang Department of Computer Science and Tecnology, Tsingua Uniersity Tsingua National Laboratory for Information Science and Tecnology

More information

THE ROLE OF U.S. TRADING IN PRICING INTERNATIONALLY CROSS-LISTED STOCKS

THE ROLE OF U.S. TRADING IN PRICING INTERNATIONALLY CROSS-LISTED STOCKS THE ROLE OF U.S. TRADING IN PRICING INTERNATIONALLY CROSS-LISTED STOCKS by Joacim Grammig a, Micael Melvin b, and Cristian Sclag c Abstract: Tis paper addresses two issues: 1) were does price discovery

More information