6. DYNAMIC PROGRAMMING II

Size: px
Start display at page:

Download "6. DYNAMIC PROGRAMMING II"

Transcription

1 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg's algorihm Bellman-Ford algorihm disance vecor proocols negaive cycles in a digraph 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg's algorihm Bellman-Ford algorihm disance vecor proocols negaive cycles in a digraph Lecure slides by Kevin Wayne Copyrigh 00 Pearson-Addison Wesley hp:// SECTION 6.6 Las updaed on Mar 6, 0 : AM Sring similariy Edi disance Q. How similar are wo srings? Ex. ocurrance and occurrence. Edi disance. [Levenshein 966, Needleman-Wunsch 970] Gap penaly δ; mismach penaly α pq. Cos = sum of gap and mismach penalies. o c u r r a n c e o c c u r r e n c e 6 mismaches, gap o c u r r a n c e o c c u r r e n c e mismach, gap C T G A C C T A C G C T G G A C G A A C G cos = δ + αcg + αta o c u r r a n c e o c c u r r e n c e 0 mismaches, gaps Applicaions. Unix diff, speech recogniion, compuaional biology,...

2 Sequence alignmen Sequence alignmen: problem srucure Goal. Given wo srings x x... x m and y y... y n find min cos alignmen. Def. OPT(i, j) = min cos of aligning prefix srings x x... x i and y y... y j. Def. An alignmen M is a se of ordered pairs x i y j such ha each iem occurs in a mos one pair and no crossings. Def. The cos of an alignmen M is: cos(m ) = α xi y j (x i, y j ) M mismach + δ + δ i : x i unmached j : y j unmached gap x i y j and x i' y j' cross if i < i ', bu j > j ' Case. OPT maches xi yj. Pay mismach for x i y j + min cos of aligning x x... x i and y y... y j. Case a. OPT leaves xi unmached. Pay gap for x i + min cos of aligning x x... x i and y y... y j. Case b. OPT leaves yj unmached. Pay gap for y j + min cos of aligning x x... x i and y y... y j. opimal subsrucure propery (proof via exchange argumen) x x x x x x 6 C T A C C G T A C A T G y y y y y y 6 an alignmen of CTACCG and TACATG: M = { x y, x y, x y, x y, x 6 y 6 } OPT(i, j) = jδ if i = 0 α xi y j +OPT(i, j ) min δ +OPT(i, j) oherwise δ +OPT(i, j ) iδ if j = 0 6 Sequence alignmen: algorihm Sequence alignmen: analysis SEQUENCE-ALIGNMENT (m, n, x,, xm, y,, yn, δ, α) FOR i = 0 TO m M [i, 0] i δ. FOR j = 0 TO n M [0, j] j δ. FOR i = TO m FOR j = TO n M [i, j] min { α[xi, yj] + M [i, j ], δ + M [i, j], δ + M [i, j ]). Theorem. The dynamic programming algorihm compues he edi disance (and opimal alignmen) of wo srings of lengh m and n in Θ(mn) ime and Θ(mn) space. Algorihm compues edi disance. Can race back o exrac opimal alignmen iself. Q. Can we avoid using quadraic space? A. Easy o compue opimal value in O(mn) ime and O(m + n) space. Compue OPT(i, ) from OPT(i, ). Bu, no longer easy o recover opimal alignmen iself. RETURN M [m, n]. 7 8

3 Sequence alignmen in linear space 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg's algorihm Bellman-Ford algorihm disance vecor proocols negaive cycles in a digraph Theorem. There exis an algorihm o find an opimal alignmen in O(mn) ime and O(m + n) space. Clever combinaion of divide-and-conquer and dynamic programming. Inspired by idea of Savich from complexiy heory. Programming Techniques G. Manacher Edior A Linear Space Algorihm for Compuing Maximal Common Subsequences D.S. Hirschberg Princeon Universiy SECTION 6.7 The problem of finding a longes common subsequence of wo srings has been solved in quadraic ime and space. An algorihm is presened which will solve his problem in quadraic ime and in linear space. Key Words and Phrases: subsequence, longes common subsequence, sring correcion, ediing CR Caegories:.6,.7,.79,.,. Inroducion 0 Hirschberg's algorihm Hirschberg's algorihm Edi disance graph. Le f (i, j) be shores pah from (0,0) o (i, j). Lemma: f (i, j) = OPT(i, j) for all i and j. Edi disance graph. Le f (i, j) be shores pah from (0,0) o (i, j). Lemma: f (i, j) = OPT(i, j) for all i and j. Pf of Lemma. [ by srong inducion on i + j ] x y y y y y y 6 Base case: f (0, 0) = OPT (0, 0) = 0. Inducive hypohesis: assume rue for all (i', j') wih i' + j' < i + j. Las edge on shores pah o (i, j) is from (i, j ), (i, j), or (i, j ). Thus, f(i, j) = min{ xiy j + f(i,j ), + f(i,j), + f(i, j )} { } α xi y j δ = min{ xiy j + OPT(i,j ), + OPT(i,j), + OPT(i, j )} { } x δ = OPT(i, j) α xi y j δ x δ

4 Hirschberg's algorihm Hirschberg's algorihm Edi disance graph. Le f (i, j) be shores pah from (0,0) o (i, j). Lemma: f (i, j) = OPT(i, j) for all i and j. Can compue f (, j) for any j in O(mn) ime and O(m + n) space. Edi disance graph. Le g (i, j) be shores pah from (i, j) o (m, n). Can compue by reversing he edge orienaions and invering he roles of (0, 0) and (m, n). j y y y y y y 6 y y y y y y 6 x x δ δ α x i y j x x x x Hirschberg's algorihm Hirschberg's algorihm Edi disance graph. Le g (i, j) be shores pah from (i, j) o (m, n). Can compue g(, j) for any j in O(mn) ime and O(m + n) space. Observaion. The cos of he shores pah ha uses (i, j) is f (i, j) + g(i, j). j y y y y y y 6 y y y y y y 6 x x x x x x 6

5 Hirschberg's algorihm Hirschberg's algorihm Observaion. le q be an index ha minimizes f (q, n/) + g (q, n/). Then, here exiss a shores pah from (0, 0) o (m, n) uses (q, n/). Divide. Find index q ha minimizes f (q, n/) + g(q, n/); align x q and y n /. Conquer. Recursively compue opimal alignmen in each piece. n / n / y y y y y y 6 y y y y y y 6 x q x q x x x x 7 8 Hirschberg's algorihm: running ime analysis warmup Hirschberg's algorihm: running ime analysis Theorem. Le T(m, n) = max running ime of Hirschberg's algorihm on srings of lengh a mos m and n. Then, T(m, n) = O(m n log n). Theorem. Le T(m, n) = max running ime of Hirschberg's algorihm on srings of lengh a mos m and n. Then, T(m, n) = O(m n). T(m, n) T(m, n / ) + O(m n) T(m, n) = O(m log n). Remark. Analysis is no igh because wo subproblems are of size (q, n/) and (m q, n / ). In nex slide, we save log n facor. [ by inducion on n ] O(m n) ime o compue f (, n / ) and g (, n / ) and find index q. T(q, n / ) + T(m q, n / ) ime for wo recursive calls. Choose consan c so ha: T(m, ) T(, n) c m c n T(m, n) c m n + T(q, n / ) + T(m q, n / ) Claim. T(m, n) c m n. Base cases: m = or n =. Inducive hypohesis: T(m, n) c m n for all (m', n') wih m' + n' < m + n. T(m, n) T(q, n / ) + T(m q, n / ) + c m n c q n / + c (m q) n / + c m n = c q n + c m n c q n + c m n = c m n 9 0

6 Shores pahs 6. DYNAMIC PROGRAMMING II Shores pah problem. Given a digraph G = (V, E), wih arbirary edge weighs or coss c vw, find cheapes pah from node s o node. sequence alignmen Hirschberg's algorihm Bellman-Ford disance vecor proocols negaive cycles in a digraph source s SECTION cos of pah = = 8 desinaion Shores pahs: failed aemps Negaive cycles Dijksra. Can fail if negaive edge weighs. s u Def. A negaive cycle is a direced cycle such ha he sum of is edge weighs is negaive. v -8 w Reweighing. Adding a consan o every edge weigh can fail. s u v w a negaive cycle W : c(w ) = - c e < 0 e W

7 Shores pahs and negaive cycles Shores pahs and negaive cycles Lemma. If some pah from v o conains a negaive cycle, hen here does no exis a cheapes pah from v o. Lemma. If G has no negaive cycles, hen here exiss a cheapes pah from v o ha is simple (and has n edges). If here exiss such a cycle W, hen can build a v pah of arbirarily negaive weigh by deouring around cycle as many imes as desired. Consider a cheapes v pah P ha uses he fewes number of edges. If P conains a cycle W, can remove porion of P corresponding o W wihou increasing he cos. v v W W c(w) < 0 c(w) 0 6 Shores pah and negaive cycle problems Shores pahs: dynamic programming Shores pah problem. Given a digraph G = (V, E) wih edge weighs c vw and no negaive cycles, find cheapes v pah for each node v. Negaive cycle problem. Given a digraph G = (V, E) wih edge weighs c vw, find a negaive cycle (if one exiss). Def. OPT(i, v) = cos of shores v pah ha uses i edges. Case : Cheapes v pah uses i edges. OPT(i, v) = OPT(i, v) Case : Cheapes v pah uses exacly i edges. if (v, w) is firs edge, hen OPT uses (v, w), and hen selecs bes w pah using i edges opimal subsrucure propery (proof via exchange argumen) - OPT(i, v) = 0 if i = 0 min OPT(i, v), min { OPT(i, w)+ c vw } (v, w) E oherwise shores-pahs ree negaive cycle Observaion. If no negaive cycles, OPT(n, v) = cos of cheapes v pah. By Lemma, cheapes v pah is simple. 7 8

8 Shores pahs: implemenaion Shores pahs: implemenaion Theorem. Given a digraph G = (V, E) wih no negaive cycles, he dynamic programming algorihm compues he cos of he cheapes v pah for SHORTEST-PATHS (V, E, c, ) each node v in Θ(mn) ime and Θ(n ) space. FOREACH node v V M [0, v]. M [0, ] 0. FOR i = TO n FOREACH node v V M [i, v] M [i, v]. FOREACH edge (v, w) E M [i, v] min { M [i, v], M [i, w] + cvw }. Table requires Θ(n ) space. Each ieraion i akes Θ(m) ime since we examine each edge once. Finding he shores pahs. Approach : Mainain a successor(i, v) ha poins o nex node on cheapes v pah using a mos i edges. Approach : Compue opimal coss M[i, v] and consider only edges wih M[i, v] = M[i, w] + c vw. 9 0 Shores pahs: pracical improvemens Bellman-Ford: efficien implemenaion Space opimizaion. Mainain wo d arrays (insead of d array). d(v) = cos of cheapes v pah ha we have found so far. successor(v) = nex node on a v pah. Performance opimizaion. If d(w) was no updaed in ieraion i, hen no reason o consider edges enering w in ieraion i. BELLMAN-FORD (V, E, c, ) FOREACH node v V d(v). successor(v) null. d() 0. FOR i = TO n FOREACH node w V IF (d(w) was updaed in previous ieraion) FOREACH edge (v, w) E IF ( d(v) > d(w) + cvw) pass d(v) d(w) + cvw. successor(v) w. IF no d(w) value changed in ieraion i, STOP.

9 Bellman-Ford: analysis Bellman-Ford: analysis Claim. Afer he i h pass of Bellman-Ford, d(v) equals he cos of he cheapes v pah using a mos i edges. Counerexample. Claim is false! d(v) = d(w) = d() = 0 Lemma. Throughou Bellman-Ford algorihm, d(v) is he cos of some v pah; afer he i h pass, d(v) is no larger han he cos of he cheapes v pah using i edges. [by inducion on i] Assume rue afer i h pass. Le P be any v pah wih i + edges. Le (v, w) be firs edge on pah and le P' be subpah from w o. By inducive hypohesis, d(w) c(p') since P' is a w pah wih i edges. Afer considering v in pass i+: d(v) c vw + d(w) cvw + c(p') = c(p) v w if nodes w considered before node v, hen d(v) = afer pass Theorem. Given a digraph wih no negaive cycles, Bellman-Ford compues he coss of he cheapes v pahs in O(mn) ime and Θ(n) exra space. Lemmas +. can be subsanially faser in pracice Bellman-Ford: analysis Bellman-Ford: analysis Claim. Throughou he Bellman-Ford algorihm, following successor(v) poiners gives a direced pah from v o of cos d(v). Claim. Throughou he Bellman-Ford algorihm, following successor(v) poiners gives a direced pah from v o of cos d(v). Counerexample. Claim is false! Cos of successor v pah may have sricly lower cos han d(v). Counerexample. Claim is false! Cos of successor v pah may have sricly lower cos han d(v). consider nodes in order:,,, consider nodes in order:,,, s() = s() = s() = s() = d() = 0 d() = 0 d() = 0 d() = 0 d() = d() = s() = d() = s() = d() = 6

10 Bellman-Ford: analysis Bellman-Ford: analysis Claim. Throughou he Bellman-Ford algorihm, following successor(v) poiners gives a direced pah from v o of cos d(v). Claim. Throughou he Bellman-Ford algorihm, following successor(v) poiners gives a direced pah from v o of cos d(v). Counerexample. Claim is false! Cos of successor v pah may have sricly lower cos han d(v). Successor graph may have cycles. Counerexample. Claim is false! Cos of successor v pah may have sricly lower cos han d(v). Successor graph may have cycles. consider nodes in order:,,,, consider nodes in order:,,,, d() = 0 d() = 8 d() = 0 d() = 8 9 d() = 0 9 d() = d() = d() = 7 d() = d() = 8 Bellman-Ford: finding he shores pah Bellman-Ford: finding he shores pah Lemma. If he successor graph conains a direced cycle W, hen W is a negaive cycle. If successor(v) = w, we mus have d(v) d(w) + cvw. (LHS and RHS are equal when successor(v) is se; d(w) can only decrease; d(v) decreases only when successor(v) is rese) Le v v vk be he nodes along he cycle W. Assume ha (v k, v ) is he las edge added o he successor graph. Jus prior o ha: d(v) d(v) + c(v, v) d(v) d(v) + c(v, v) d(vk ) d(vk) + c(vk, vk) d(vk) > d(v) + c(vk, v) Adding inequaliies yields c(v, v) + c(v, v) + + c(vk, vk) + c(vk, v) < 0. W is a negaive cycle holds wih sric inequaliy since we are updaing d(vk) 9 Theorem. Given a digraph wih no negaive cycles, Bellman-Ford finds he cheapes s pahs in O(mn) ime and Θ(n) exra space. The successor graph canno have a negaive cycle. [Lemma ] Thus, following he successor poiners from s yields a direced pah o. Le s = v v vk = be he nodes along his pah P. Upon erminaion, if successor(v) = w, we mus have d(v) = d(w) + c vw. (LHS and RHS are equal when successor(v) is se; d( ) did no change) Thus, d(v) = d(v) + c(v, v) d(v) = d(v) + c(v, v) d(vk ) = d(vk) + c(vk, vk) Adding equaions yields d(s) = d() + c(v, v) + c(v, v) + + c(vk, vk). min cos of any s pah (Theorem ) 0 cos of pah P since algorihm erminaed 0

11 Disance vecor proocols SECTION DYNAMIC PROGRAMMING II sequence alignmen Hirschberg's algorihm Bellman-Ford disance vecor proocols negaive cycles in a digraph Communicaion nework. Node rouer. Edge direc communicaion link. Cos of edge delay on link. Dijksra's algorihm. Requires global informaion of nework. Bellman-Ford. Uses only local knowledge of neighboring nodes. Synchronizaion. We don' expec rouers o run in locksep. The order in which each foreach loop execues in no imporan. Moreover, algorihm sill converges even if updaes are asynchronous. naurally nonnegaive, bu Bellman-Ford used anyway! Disance vecor proocols Pah vecor proocols Disance vecor proocols. [ "rouing by rumor" ] Each rouer mainains a vecor of shores pah lenghs o every oher node (disances) and he firs hop on each pah (direcions). Algorihm: each rouer performs n separae compuaions, one for each poenial desinaion node. Ex. RIP, Xerox XNS RIP, Novell's IPX RIP, Cisco's IGRP, DEC's DNA Phase IV, AppleTalk's RTMP. Link sae rouing. Each rouer also sores he enire pah. Based on Dijksra's algorihm. Avoids "couning-o-infiniy" problem and relaed difficulies. Requires significanly more sorage. no jus he disance and firs hop Ex. Border Gaeway Proocol (BGP), Open Shores Pah Firs (OSPF). Cavea. Edge coss may change during algorihm (or fail compleely). deleed s v d(s) = d(v) = d() = 0 "couning o infiniy"

12 Deecing negaive cycles 6. DYNAMIC PROGRAMMING II Negaive cycle deecion problem. Given a digraph G = (V, E), wih edge weighs c vw, find a negaive cycle (if one exiss). sequence alignmen Hirschberg's algorihm Bellman-Ford disance vecor proocol negaive cycles in a digraph 6 SECTION Deecing negaive cycles: applicaion Deecing negaive cycles Currency conversion. Given n currencies and exchange raes beween pairs of currencies, is here an arbirage opporuniy? Remark. Fases algorihm very valuable! 0.7 *.66 *.99 = EUR USD GBP Lemma. If OPT(n, v) = OPT(n, v) for all v, hen no negaive cycle can reach. Bellman-Ford algorihm. Lemma 6. If OPT(n, v) < OPT(n, v) for some node v, hen (any) cheapes pah from v o conains a cycle W. Moreover W is a negaive cycle. [by conradicion] Since OPT(n, v) < OPT(n, v), we know ha shores v pah P has exacly n edges. By pigeonhole principle, P mus conain a direced cycle W. Deleing W yields a v pah wih < n edges W has negaive cos. v W CAD 0.9 CHF 7 c(w) < 0 8

13 Deecing negaive cycles Deecing negaive cycles Theorem. Can find a negaive cycle in Θ(mn) ime and Θ(n ) space. Add new node and connec all nodes o wih 0-cos edge. G has a negaive cycle iff G' has a negaive cycle han can reach. If OPT(n, v) = OPT(n, v) for all nodes v, hen no negaive cycles. If no, hen exrac direced cycle from pah from v o. (cycle canno conain since no edges leave ) Theorem. Can find a negaive cycle in O(mn) ime and O(n) exra space. Run Bellman-Ford for n passes (insead of n ) on modified digraph. If no d(v) values updaed in pass n, hen no negaive cycles. Oherwise, suppose d(s) updaed in pass n. Define pass(v) = las pass in which d(v) was updaed. Observe pass(s) = n and pass(successor(v)) pass(v) for each v. Following successor poiners, we mus evenually repea a node. Lemma his cycle is a negaive cycle. 6 G' Remark. See p. 0 for improved version and early erminaion rule. (Tarjan's subree disassembly rick) - 9 0

6. DYNAMIC PROGRAMMING II

6. DYNAMIC PROGRAMMING II 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg s algorihm Bellman Ford algorihm disance vecor proocols negaive cycles in a digraph 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg s algorihm

More information

6. DYNAMIC PROGRAMMING II

6. DYNAMIC PROGRAMMING II 6. DYNAMIC PROGRAMMING II sequence alignment Hirschberg's algorithm Bellman-Ford algorithm distance vector protocols negative cycles in a digraph Lecture slides by Kevin Wayne Copyright 2005 Pearson-Addison

More information

6. DYNAMIC PROGRAMMING II

6. DYNAMIC PROGRAMMING II 6. DYNAMIC PROGRAMMING II 6. DYNAMIC PROGRAMMING II sequence alignmen Hirschberg s algorihm Bellman Ford Moore algorihm disance-vecor proocols negaive cycles sequence alignmen Hirschberg s algorihm Bellman

More information

6. DYNAMIC PROGRAMMING II. sequence alignment Hirschberg's algorithm Bellman-Ford distance vector protocols negative cycles in a digraph

6. DYNAMIC PROGRAMMING II. sequence alignment Hirschberg's algorithm Bellman-Ford distance vector protocols negative cycles in a digraph 6. DYNAMIC PROGRAMMING II sequence alignment Hirschberg's algorithm Bellman-Ford distance vector protocols negative cycles in a digraph Shortest paths Shortest path problem. Given a digraph G = (V, E),

More information

Chapter 6. Dynamic Programming. Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved.

Chapter 6. Dynamic Programming. Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved. Chapter 6 Dynamic Programming Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved. 1 6.8 Shortest Paths Shortest Paths Shortest path problem. Given a directed graph G = (V,

More information

Dynamic Programming 11/8/2009. Weighted Interval Scheduling. Weighted Interval Scheduling. Unweighted Interval Scheduling: Review

Dynamic Programming 11/8/2009. Weighted Interval Scheduling. Weighted Interval Scheduling. Unweighted Interval Scheduling: Review //9 Algorihms Dynamic Programming - Weighed Ineral Scheduling Dynamic Programming Weighed ineral scheduling problem. Insance A se of n jobs. Job j sars a s j, finishes a f j, and has weigh or alue j. Two

More information

Data Structures and Algorithms CMPSC 465

Data Structures and Algorithms CMPSC 465 ata Structures and Algorithms MPS LTUR ellman-ford Shortest Paths Algorithm etecting Negative ost ycles Adam Smith // A. Smith; based on slides by K. Wayne,. Leiserson,. emaine L. Negative Weight dges

More information

6.6 Sequence Alignment

6.6 Sequence Alignment 6.6 Sequence Alignment String Similarity How similar are two strings? ocurrance o c u r r a n c e - occurrence First model the problem Q. How can we measure the distance? o c c u r r e n c e 6 mismatches,

More information

CSE 202 Dynamic Programming II

CSE 202 Dynamic Programming II CSE 202 Dynamic Programming II Chapter 6 Dynamic Programming Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved. 1 Algorithmic Paradigms Greed. Build up a solution incrementally,

More information

Chapter 6. Dynamic Programming. Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved.

Chapter 6. Dynamic Programming. Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved. Chapter 6 Dynamic Programming Slides by Kevin Wayne. Copyright 2005 Pearson-Addison Wesley. All rights reserved. 1 Algorithmic Paradigms Greed. Build up a solution incrementally, myopically optimizing

More information

Longest Common Prefixes

Longest Common Prefixes Longes Common Prefixes The sandard ordering for srings is he lexicographical order. I is induced by an order over he alphabe. We will use he same symbols (,

More information

CS 580: Algorithm Design and Analysis

CS 580: Algorithm Design and Analysis CS 58: Algorithm Design and Analysis Jeremiah Blocki Purdue University Spring 28 Announcement: Homework 3 due February 5 th at :59PM Midterm Exam: Wed, Feb 2 (8PM-PM) @ MTHW 2 Recap: Dynamic Programming

More information

A Hop Constrained Min-Sum Arborescence with Outage Costs

A Hop Constrained Min-Sum Arborescence with Outage Costs A Hop Consrained Min-Sum Arborescence wih Ouage Coss Rakesh Kawara Minnesoa Sae Universiy, Mankao, MN 56001 Email: Kawara@mnsu.edu Absrac The hop consrained min-sum arborescence wih ouage coss problem

More information

The Residual Graph. 11 Augmenting Path Algorithms. Augmenting Path Algorithm. Augmenting Path Algorithm

The Residual Graph. 11 Augmenting Path Algorithms. Augmenting Path Algorithm. Augmenting Path Algorithm Augmening Pah Algorihm Greedy-algorihm: ar wih f (e) = everywhere find an - pah wih f (e) < c(e) on every edge augmen flow along he pah repea a long a poible The Reidual Graph From he graph G = (V, E,

More information

The Residual Graph. 12 Augmenting Path Algorithms. Augmenting Path Algorithm. Augmenting Path Algorithm

The Residual Graph. 12 Augmenting Path Algorithms. Augmenting Path Algorithm. Augmenting Path Algorithm Augmening Pah Algorihm Greedy-algorihm: ar wih f (e) = everywhere find an - pah wih f (e) < c(e) on every edge augmen flow along he pah repea a long a poible The Reidual Graph From he graph G = (V, E,

More information

1 Review of Zero-Sum Games

1 Review of Zero-Sum Games COS 5: heoreical Machine Learning Lecurer: Rob Schapire Lecure #23 Scribe: Eugene Brevdo April 30, 2008 Review of Zero-Sum Games Las ime we inroduced a mahemaical model for wo player zero-sum games. Any

More information

16 Max-Flow Algorithms

16 Max-Flow Algorithms A process canno be undersood by sopping i. Undersanding mus move wih he flow of he process, mus join i and flow wih i. The Firs Law of Mena, in Frank Herber s Dune (196) There s a difference beween knowing

More information

Random Walk with Anti-Correlated Steps

Random Walk with Anti-Correlated Steps Random Walk wih Ani-Correlaed Seps John Noga Dirk Wagner 2 Absrac We conjecure he expeced value of random walks wih ani-correlaed seps o be exacly. We suppor his conjecure wih 2 plausibiliy argumens and

More information

Traversal of a subtree is slow, which affects prefix and range queries.

Traversal of a subtree is slow, which affects prefix and range queries. Compac Tries Tries suffer from a large number nodes, Ω( R ) in he wors case. The space requiremen is large, since each node needs much more space han a single symbol. Traversal of a subree is slow, which

More information

Lecture 2-1 Kinematics in One Dimension Displacement, Velocity and Acceleration Everything in the world is moving. Nothing stays still.

Lecture 2-1 Kinematics in One Dimension Displacement, Velocity and Acceleration Everything in the world is moving. Nothing stays still. Lecure - Kinemaics in One Dimension Displacemen, Velociy and Acceleraion Everyhing in he world is moving. Nohing says sill. Moion occurs a all scales of he universe, saring from he moion of elecrons in

More information

EXERCISES FOR SECTION 1.5

EXERCISES FOR SECTION 1.5 1.5 Exisence and Uniqueness of Soluions 43 20. 1 v c 21. 1 v c 1 2 4 6 8 10 1 2 2 4 6 8 10 Graph of approximae soluion obained using Euler s mehod wih = 0.1. Graph of approximae soluion obained using Euler

More information

Max Flow, Min Cut COS 521. Kevin Wayne Fall Soviet Rail Network, Cuts. Minimum Cut Problem. Flow network.

Max Flow, Min Cut COS 521. Kevin Wayne Fall Soviet Rail Network, Cuts. Minimum Cut Problem. Flow network. Sovie Rail Nework, Max Flow, Min u OS Kevin Wayne Fall Reference: On he hiory of he ranporaion and maximum flow problem. lexander Schrijver in Mah Programming, :,. Minimum u Problem u Flow nework.! Digraph

More information

Instructor: Barry McQuarrie Page 1 of 5

Instructor: Barry McQuarrie Page 1 of 5 Procedure for Solving radical equaions 1. Algebraically isolae one radical by iself on one side of equal sign. 2. Raise each side of he equaion o an appropriae power o remove he radical. 3. Simplify. 4.

More information

MATH 5720: Gradient Methods Hung Phan, UMass Lowell October 4, 2018

MATH 5720: Gradient Methods Hung Phan, UMass Lowell October 4, 2018 MATH 5720: Gradien Mehods Hung Phan, UMass Lowell Ocober 4, 208 Descen Direcion Mehods Consider he problem min { f(x) x R n}. The general descen direcions mehod is x k+ = x k + k d k where x k is he curren

More information

CMU-Q Lecture 3: Search algorithms: Informed. Teacher: Gianni A. Di Caro

CMU-Q Lecture 3: Search algorithms: Informed. Teacher: Gianni A. Di Caro CMU-Q 5-38 Lecure 3: Search algorihms: Informed Teacher: Gianni A. Di Caro UNINFORMED VS. INFORMED SEARCH Sraegy How desirable is o be in a cerain inermediae sae for he sake of (effecively) reaching a

More information

Dynamic Programming. Weighted Interval Scheduling. Algorithmic Paradigms. Dynamic Programming

Dynamic Programming. Weighted Interval Scheduling. Algorithmic Paradigms. Dynamic Programming lgorithmic Paradigms Dynamic Programming reed Build up a solution incrementally, myopically optimizing some local criterion Divide-and-conquer Break up a problem into two sub-problems, solve each sub-problem

More information

Stationary Distribution. Design and Analysis of Algorithms Andrei Bulatov

Stationary Distribution. Design and Analysis of Algorithms Andrei Bulatov Saionary Disribuion Design and Analysis of Algorihms Andrei Bulaov Algorihms Markov Chains 34-2 Classificaion of Saes k By P we denoe he (i,j)-enry of i, j Sae is accessible from sae if 0 for some k 0

More information

A Primal-Dual Type Algorithm with the O(1/t) Convergence Rate for Large Scale Constrained Convex Programs

A Primal-Dual Type Algorithm with the O(1/t) Convergence Rate for Large Scale Constrained Convex Programs PROC. IEEE CONFERENCE ON DECISION AND CONTROL, 06 A Primal-Dual Type Algorihm wih he O(/) Convergence Rae for Large Scale Consrained Convex Programs Hao Yu and Michael J. Neely Absrac This paper considers

More information

Online Convex Optimization Example And Follow-The-Leader

Online Convex Optimization Example And Follow-The-Leader CSE599s, Spring 2014, Online Learning Lecure 2-04/03/2014 Online Convex Opimizaion Example And Follow-The-Leader Lecurer: Brendan McMahan Scribe: Sephen Joe Jonany 1 Review of Online Convex Opimizaion

More information

Stability. Coefficients may change over time. Evolution of the economy Policy changes

Stability. Coefficients may change over time. Evolution of the economy Policy changes Sabiliy Coefficiens may change over ime Evoluion of he economy Policy changes Time Varying Parameers y = α + x β + Coefficiens depend on he ime period If he coefficiens vary randomly and are unpredicable,

More information

Soviet Rail Network, 1955

Soviet Rail Network, 1955 Sovie Rail Nework, 1 Reference: On he hiory of he ranporaion and maximum flow problem. Alexander Schrijver in Mah Programming, 1: 3,. Maximum Flow and Minimum Cu Max flow and min cu. Two very rich algorihmic

More information

Timed Circuits. Asynchronous Circuit Design. Timing Relationships. A Simple Example. Timed States. Timing Sequences. ({r 6 },t6 = 1.

Timed Circuits. Asynchronous Circuit Design. Timing Relationships. A Simple Example. Timed States. Timing Sequences. ({r 6 },t6 = 1. Timed Circuis Asynchronous Circui Design Chris J. Myers Lecure 7: Timed Circuis Chaper 7 Previous mehods only use limied knowledge of delays. Very robus sysems, bu exremely conservaive. Large funcional

More information

cs/ee 143 Communication Networks

cs/ee 143 Communication Networks cs/ee 143 Communicaion Neworks Chaper 3 Eherne Tex: Walrand & Parakh, 2010 Seven Low CMS, EE, Calech Warning These noes are no self-conained, probably no undersandable, unless you also were in he lecure

More information

Extending the Four Russian Algorithm to Compute the Edit Script in Linear Space. Abstract

Extending the Four Russian Algorithm to Compute the Edit Script in Linear Space. Abstract Exing he Four Russian Algorihm o Compue he Edi Scrip in Linear Space Vamsi Kundei (vamsik@engr.uconn.edu) CSE Deparmen Universiy of Connecicu Sorrs,CT Sanguhevar Rajasekaran (rajasek@engr.uconn.edu) CSE

More information

Essential Microeconomics : OPTIMAL CONTROL 1. Consider the following class of optimization problems

Essential Microeconomics : OPTIMAL CONTROL 1. Consider the following class of optimization problems Essenial Microeconomics -- 6.5: OPIMAL CONROL Consider he following class of opimizaion problems Max{ U( k, x) + U+ ( k+ ) k+ k F( k, x)}. { x, k+ } = In he language of conrol heory, he vecor k is he vecor

More information

Randomized Perfect Bipartite Matching

Randomized Perfect Bipartite Matching Inenive Algorihm Lecure 24 Randomized Perfec Biparie Maching Lecurer: Daniel A. Spielman April 9, 208 24. Inroducion We explain a randomized algorihm by Ahih Goel, Michael Kapralov and Sanjeev Khanna for

More information

Technical Report Doc ID: TR March-2013 (Last revision: 23-February-2016) On formulating quadratic functions in optimization models.

Technical Report Doc ID: TR March-2013 (Last revision: 23-February-2016) On formulating quadratic functions in optimization models. Technical Repor Doc ID: TR--203 06-March-203 (Las revision: 23-Februar-206) On formulaing quadraic funcions in opimizaion models. Auhor: Erling D. Andersen Convex quadraic consrains quie frequenl appear

More information

! Abstraction for material flowing through the edges. ! G = (V, E) = directed graph, no parallel edges.

! Abstraction for material flowing through the edges. ! G = (V, E) = directed graph, no parallel edges. Sovie Rail Nework, haper Nework Flow Slide by Kevin Wayne. opyrigh Pearon-ddion Weley. ll righ reerved. Reference: On he hiory of he ranporaion and maximum flow problem. lexander Schrijver in Mah Programming,

More information

4 Sequences of measurable functions

4 Sequences of measurable functions 4 Sequences of measurable funcions 1. Le (Ω, A, µ) be a measure space (complee, afer a possible applicaion of he compleion heorem). In his chaper we invesigae relaions beween various (nonequivalen) convergences

More information

T L. t=1. Proof of Lemma 1. Using the marginal cost accounting in Equation(4) and standard arguments. t )+Π RB. t )+K 1(Q RB

T L. t=1. Proof of Lemma 1. Using the marginal cost accounting in Equation(4) and standard arguments. t )+Π RB. t )+K 1(Q RB Elecronic Companion EC.1. Proofs of Technical Lemmas and Theorems LEMMA 1. Le C(RB) be he oal cos incurred by he RB policy. Then we have, T L E[C(RB)] 3 E[Z RB ]. (EC.1) Proof of Lemma 1. Using he marginal

More information

Non-parametric techniques. Instance Based Learning. NN Decision Boundaries. Nearest Neighbor Algorithm. Distance metric important

Non-parametric techniques. Instance Based Learning. NN Decision Boundaries. Nearest Neighbor Algorithm. Distance metric important on-parameric echniques Insance Based Learning AKA: neares neighbor mehods, non-parameric, lazy, memorybased, or case-based learning Copyrigh 2005 by David Helmbold 1 Do no fi a model (as do LDA, logisic

More information

Math 10B: Mock Mid II. April 13, 2016

Math 10B: Mock Mid II. April 13, 2016 Name: Soluions Mah 10B: Mock Mid II April 13, 016 1. ( poins) Sae, wih jusificaion, wheher he following saemens are rue or false. (a) If a 3 3 marix A saisfies A 3 A = 0, hen i canno be inverible. True.

More information

Notes for Lecture 17-18

Notes for Lecture 17-18 U.C. Berkeley CS278: Compuaional Complexiy Handou N7-8 Professor Luca Trevisan April 3-8, 2008 Noes for Lecure 7-8 In hese wo lecures we prove he firs half of he PCP Theorem, he Amplificaion Lemma, up

More information

Lecture 20: Riccati Equations and Least Squares Feedback Control

Lecture 20: Riccati Equations and Least Squares Feedback Control 34-5 LINEAR SYSTEMS Lecure : Riccai Equaions and Leas Squares Feedback Conrol 5.6.4 Sae Feedback via Riccai Equaions A recursive approach in generaing he marix-valued funcion W ( ) equaion for i for he

More information

Designing Information Devices and Systems I Spring 2019 Lecture Notes Note 17

Designing Information Devices and Systems I Spring 2019 Lecture Notes Note 17 EES 16A Designing Informaion Devices and Sysems I Spring 019 Lecure Noes Noe 17 17.1 apaciive ouchscreen In he las noe, we saw ha a capacior consiss of wo pieces on conducive maerial separaed by a nonconducive

More information

From Particles to Rigid Bodies

From Particles to Rigid Bodies Rigid Body Dynamics From Paricles o Rigid Bodies Paricles No roaions Linear velociy v only Rigid bodies Body roaions Linear velociy v Angular velociy ω Rigid Bodies Rigid bodies have boh a posiion and

More information

Chapter 6. Weighted Interval Scheduling. Dynamic Programming. Algorithmic Paradigms. Dynamic Programming Applications

Chapter 6. Weighted Interval Scheduling. Dynamic Programming. Algorithmic Paradigms. Dynamic Programming Applications lgorithmic Paradigms hapter Dynamic Programming reedy. Build up a solution incrementally, myopically optimizing some local criterion. Divide-and-conquer. Break up a problem into sub-problems, solve each

More information

Maximum Flow in Planar Graphs

Maximum Flow in Planar Graphs Maximum Flow in Planar Graph Planar Graph and i Dual Dualiy i defined for direced planar graph a well Minimum - cu in undireced planar graph An - cu (undireced graph) An - cu The dual o he cu Cu/Cycle

More information

Vehicle Arrival Models : Headway

Vehicle Arrival Models : Headway Chaper 12 Vehicle Arrival Models : Headway 12.1 Inroducion Modelling arrival of vehicle a secion of road is an imporan sep in raffic flow modelling. I has imporan applicaion in raffic flow simulaion where

More information

PHYSICS 220 Lecture 02 Motion, Forces, and Newton s Laws Textbook Sections

PHYSICS 220 Lecture 02 Motion, Forces, and Newton s Laws Textbook Sections PHYSICS 220 Lecure 02 Moion, Forces, and Newon s Laws Texbook Secions 2.2-2.4 Lecure 2 Purdue Universiy, Physics 220 1 Overview Las Lecure Unis Scienific Noaion Significan Figures Moion Displacemen: Δx

More information

Christos Papadimitriou & Luca Trevisan November 22, 2016

Christos Papadimitriou & Luca Trevisan November 22, 2016 U.C. Bereley CS170: Algorihms Handou LN-11-22 Chrisos Papadimiriou & Luca Trevisan November 22, 2016 Sreaming algorihms In his lecure and he nex one we sudy memory-efficien algorihms ha process a sream

More information

We just finished the Erdős-Stone Theorem, and ex(n, F ) (1 1/(χ(F ) 1)) ( n

We just finished the Erdős-Stone Theorem, and ex(n, F ) (1 1/(χ(F ) 1)) ( n Lecure 3 - Kövari-Sós-Turán Theorem Jacques Versraëe jacques@ucsd.edu We jus finished he Erdős-Sone Theorem, and ex(n, F ) ( /(χ(f ) )) ( n 2). So we have asympoics when χ(f ) 3 bu no when χ(f ) = 2 i.e.

More information

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation Course Noes for EE7C Spring 018: Convex Opimizaion and Approximaion Insrucor: Moriz Hard Email: hard+ee7c@berkeley.edu Graduae Insrucor: Max Simchowiz Email: msimchow+ee7c@berkeley.edu Ocober 15, 018 3

More information

I. Introduction to place/transition nets. Place/Transition Nets I. Example: a vending machine. Example: a vending machine

I. Introduction to place/transition nets. Place/Transition Nets I. Example: a vending machine. Example: a vending machine Inroducory Tuorial I. Inroducion o place/ransiion nes Place/Transiion Nes I Prepared by: Jörg Desel, Caholic Universiy in Eichsä and Karsen Schmid, Humbold-Universiä zu Berlin Speaker: Wolfgang Reisig,

More information

Chapter 6. Dynamic Programming. CS 350: Winter 2018

Chapter 6. Dynamic Programming. CS 350: Winter 2018 Chapter 6 Dynamic Programming CS 350: Winter 2018 1 Algorithmic Paradigms Greedy. Build up a solution incrementally, myopically optimizing some local criterion. Divide-and-conquer. Break up a problem into

More information

On Boundedness of Q-Learning Iterates for Stochastic Shortest Path Problems

On Boundedness of Q-Learning Iterates for Stochastic Shortest Path Problems MATHEMATICS OF OPERATIONS RESEARCH Vol. 38, No. 2, May 2013, pp. 209 227 ISSN 0364-765X (prin) ISSN 1526-5471 (online) hp://dx.doi.org/10.1287/moor.1120.0562 2013 INFORMS On Boundedness of Q-Learning Ieraes

More information

Network Flow. Data Structures and Algorithms Andrei Bulatov

Network Flow. Data Structures and Algorithms Andrei Bulatov Nework Flow Daa Srucure and Algorihm Andrei Bulao Algorihm Nework Flow 24-2 Flow Nework Think of a graph a yem of pipe We ue hi yem o pump waer from he ource o ink Eery pipe/edge ha limied capaciy Flow

More information

Estimation of Poses with Particle Filters

Estimation of Poses with Particle Filters Esimaion of Poses wih Paricle Filers Dr.-Ing. Bernd Ludwig Chair for Arificial Inelligence Deparmen of Compuer Science Friedrich-Alexander-Universiä Erlangen-Nürnberg 12/05/2008 Dr.-Ing. Bernd Ludwig (FAU

More information

Robot Motion Model EKF based Localization EKF SLAM Graph SLAM

Robot Motion Model EKF based Localization EKF SLAM Graph SLAM Robo Moion Model EKF based Localizaion EKF SLAM Graph SLAM General Robo Moion Model Robo sae v r Conrol a ime Sae updae model Noise model of robo conrol Noise model of conrol Robo moion model

More information

Non-parametric techniques. Instance Based Learning. NN Decision Boundaries. Nearest Neighbor Algorithm. Distance metric important

Non-parametric techniques. Instance Based Learning. NN Decision Boundaries. Nearest Neighbor Algorithm. Distance metric important on-parameric echniques Insance Based Learning AKA: neares neighbor mehods, non-parameric, lazy, memorybased, or case-based learning Copyrigh 2005 by David Helmbold 1 Do no fi a model (as do LTU, decision

More information

Some Ramsey results for the n-cube

Some Ramsey results for the n-cube Some Ramsey resuls for he n-cube Ron Graham Universiy of California, San Diego Jozsef Solymosi Universiy of Briish Columbia, Vancouver, Canada Absrac In his noe we esablish a Ramsey-ype resul for cerain

More information

Supplement for Stochastic Convex Optimization: Faster Local Growth Implies Faster Global Convergence

Supplement for Stochastic Convex Optimization: Faster Local Growth Implies Faster Global Convergence Supplemen for Sochasic Convex Opimizaion: Faser Local Growh Implies Faser Global Convergence Yi Xu Qihang Lin ianbao Yang Proof of heorem heorem Suppose Assumpion holds and F (w) obeys he LGC (6) Given

More information

Notes on online convex optimization

Notes on online convex optimization Noes on online convex opimizaion Karl Sraos Online convex opimizaion (OCO) is a principled framework for online learning: OnlineConvexOpimizaion Inpu: convex se S, number of seps T For =, 2,..., T : Selec

More information

Optimal Embedding of Functions for In-Network Computation: Complexity Analysis and Algorithms

Optimal Embedding of Functions for In-Network Computation: Complexity Analysis and Algorithms 1 Opimal Embedding of Funcions for In-Nework Compuaion: Complexiy Analysis and Algorihms Pooja Vyavahare, Nuan Limaye, and D. Manjunah, Senior Member, IEEE arxiv:1401.2518v3 [cs.dc] 14 Jul 2015 Absrac

More information

Matrix Versions of Some Refinements of the Arithmetic-Geometric Mean Inequality

Matrix Versions of Some Refinements of the Arithmetic-Geometric Mean Inequality Marix Versions of Some Refinemens of he Arihmeic-Geomeric Mean Inequaliy Bao Qi Feng and Andrew Tonge Absrac. We esablish marix versions of refinemens due o Alzer ], Carwrigh and Field 4], and Mercer 5]

More information

Homework sheet Exercises done during the lecture of March 12, 2014

Homework sheet Exercises done during the lecture of March 12, 2014 EXERCISE SESSION 2A FOR THE COURSE GÉOMÉTRIE EUCLIDIENNE, NON EUCLIDIENNE ET PROJECTIVE MATTEO TOMMASINI Homework shee 3-4 - Exercises done during he lecure of March 2, 204 Exercise 2 Is i rue ha he parameerized

More information

GOAL: A Parsimonious Geographic Routing Protocol for Large Scale Sensor Networks

GOAL: A Parsimonious Geographic Routing Protocol for Large Scale Sensor Networks TRANSACTIONS ON PARALLEL AND DISTRIBUTED COMPUTING, VOL. X, NO. X, SEPTEMBER 2 : A Parsimonious Geographic Rouing Proocol for Large Scale Sensor Neworks Myounggyu Won, Wei Zhang and Radu Soleru Absrac

More information

Solutions for Assignment 2

Solutions for Assignment 2 Faculy of rs and Science Universiy of Torono CSC 358 - Inroducion o Compuer Neworks, Winer 218 Soluions for ssignmen 2 Quesion 1 (2 Poins): Go-ack n RQ In his quesion, we review how Go-ack n RQ can be

More information

L07. KALMAN FILTERING FOR NON-LINEAR SYSTEMS. NA568 Mobile Robotics: Methods & Algorithms

L07. KALMAN FILTERING FOR NON-LINEAR SYSTEMS. NA568 Mobile Robotics: Methods & Algorithms L07. KALMAN FILTERING FOR NON-LINEAR SYSTEMS NA568 Mobile Roboics: Mehods & Algorihms Today s Topic Quick review on (Linear) Kalman Filer Kalman Filering for Non-Linear Sysems Exended Kalman Filer (EKF)

More information

Algorithm Design and Analysis

Algorithm Design and Analysis Algorihm Deign and Analyi LECTURE 0 Nework Flow Applicaion Biparie maching Edge-dijoin pah Adam Smih 0//0 A. Smih; baed on lide by E. Demaine, C. Leieron, S. Rakhodnikova, K. Wayne La ime: Ford-Fulkeron

More information

THE MATRIX-TREE THEOREM

THE MATRIX-TREE THEOREM THE MATRIX-TREE THEOREM 1 The Marix-Tree Theorem. The Marix-Tree Theorem is a formula for he number of spanning rees of a graph in erms of he deerminan of a cerain marix. We begin wih he necessary graph-heoreical

More information

The Characterization of Data-Accumulating Algorithms

The Characterization of Data-Accumulating Algorithms The Characerizaion of Daa-Accumulaing Algorihms Sefan D Bruda and Selim G Akl Deparmen of Compuing and Informaion Science Queen s Universiy Kingson, Onario, K7L 3N6 Canada fbruda,aklg@csqueensuca Absrac

More information

Objec&ves. Review. Dynamic Programming. What is the knapsack problem? What is our solu&on? Ø Review Knapsack Ø Sequence Alignment 3/28/18

Objec&ves. Review. Dynamic Programming. What is the knapsack problem? What is our solu&on? Ø Review Knapsack Ø Sequence Alignment 3/28/18 /8/8 Objec&ves Dynamic Programming Ø Review Knapsack Ø Sequence Alignment Mar 8, 8 CSCI - Sprenkle Review What is the knapsack problem? What is our solu&on? Mar 8, 8 CSCI - Sprenkle /8/8 Dynamic Programming:

More information

Ensamble methods: Bagging and Boosting

Ensamble methods: Bagging and Boosting Lecure 21 Ensamble mehods: Bagging and Boosing Milos Hauskrech milos@cs.pi.edu 5329 Senno Square Ensemble mehods Mixure of expers Muliple base models (classifiers, regressors), each covers a differen par

More information

Stopping Set Elimination for LDPC Codes

Stopping Set Elimination for LDPC Codes Sopping Se Eliminaion for LDPC Codes Anxiao (Andrew) Jiang, Pulakesh Upadhyaya, Ying Wang, Krishna R. Narayanan Hongchao Zhou, Jin Sima, and Jehoshua Bruck Compuer Science and Engineering Deparmen, Texas

More information

Viterbi Algorithm: Background

Viterbi Algorithm: Background Vierbi Algorihm: Background Jean Mark Gawron March 24, 2014 1 The Key propery of an HMM Wha is an HMM. Formally, i has he following ingrediens: 1. a se of saes: S 2. a se of final saes: F 3. an iniial

More information

Article from. Predictive Analytics and Futurism. July 2016 Issue 13

Article from. Predictive Analytics and Futurism. July 2016 Issue 13 Aricle from Predicive Analyics and Fuurism July 6 Issue An Inroducion o Incremenal Learning By Qiang Wu and Dave Snell Machine learning provides useful ools for predicive analyics The ypical machine learning

More information

Algorithmic Discrete Mathematics 6. Exercise Sheet

Algorithmic Discrete Mathematics 6. Exercise Sheet Algorihmic Dicree Mahemaic. Exercie Shee Deparmen of Mahemaic SS 0 PD Dr. Ulf Lorenz 7. and 8. Juni 0 Dipl.-Mah. David Meffer Verion of June, 0 Groupwork Exercie G (Heap-Sor) Ue Heap-Sor wih a min-heap

More information

Approximate String Matching. Department of Computer Science. University of Chile. Blanco Encalada Santiago - Chile

Approximate String Matching. Department of Computer Science. University of Chile. Blanco Encalada Santiago - Chile Very Fas and Simple Approximae Sring Maching Gonzalo Navarro Ricardo Baeza-Yaes Deparmen of Compuer Science Universiy of Chile Blanco Encalada 2120 - Saniago - Chile fgnavarro,rbaezag@dcc.uchile.cl Absrac

More information

Internet Traffic Modeling for Efficient Network Research Management Prof. Zhili Sun, UniS Zhiyong Liu, CATR

Internet Traffic Modeling for Efficient Network Research Management Prof. Zhili Sun, UniS Zhiyong Liu, CATR Inerne Traffic Modeling for Efficien Nework Research Managemen Prof. Zhili Sun, UniS Zhiyong Liu, CATR UK-China Science Bridge Workshop 13-14 December 2011, London Ouline Inroducion Background Classical

More information

Matching. Slides designed by Kevin Wayne.

Matching. Slides designed by Kevin Wayne. Maching Maching. Inpu: undireced graph G = (V, E). M E i a maching if each node appear in a mo edge in M. Max maching: find a max cardinaliy maching. Slide deigned by Kevin Wayne. Biparie Maching Biparie

More information

3.1.3 INTRODUCTION TO DYNAMIC OPTIMIZATION: DISCRETE TIME PROBLEMS. A. The Hamiltonian and First-Order Conditions in a Finite Time Horizon

3.1.3 INTRODUCTION TO DYNAMIC OPTIMIZATION: DISCRETE TIME PROBLEMS. A. The Hamiltonian and First-Order Conditions in a Finite Time Horizon 3..3 INRODUCION O DYNAMIC OPIMIZAION: DISCREE IME PROBLEMS A. he Hamilonian and Firs-Order Condiions in a Finie ime Horizon Define a new funcion, he Hamilonian funcion, H. H he change in he oal value of

More information

Ensamble methods: Boosting

Ensamble methods: Boosting Lecure 21 Ensamble mehods: Boosing Milos Hauskrech milos@cs.pi.edu 5329 Senno Square Schedule Final exam: April 18: 1:00-2:15pm, in-class Term projecs April 23 & April 25: a 1:00-2:30pm in CS seminar room

More information

Areas. ! Bioinformatics. ! Control theory. ! Information theory. ! Operations research. ! Computer science: theory, graphics, AI, systems,.

Areas. ! Bioinformatics. ! Control theory. ! Information theory. ! Operations research. ! Computer science: theory, graphics, AI, systems,. lgorithmic Paradigms hapter Dynamic Programming reed Build up a solution incrementally, myopically optimizing some local criterion Divide-and-conquer Break up a problem into two sub-problems, solve each

More information

An Introduction to Malliavin calculus and its applications

An Introduction to Malliavin calculus and its applications An Inroducion o Malliavin calculus and is applicaions Lecure 5: Smoohness of he densiy and Hörmander s heorem David Nualar Deparmen of Mahemaics Kansas Universiy Universiy of Wyoming Summer School 214

More information

Adaptation and Synchronization over a Network: stabilization without a reference model

Adaptation and Synchronization over a Network: stabilization without a reference model Adapaion and Synchronizaion over a Nework: sabilizaion wihou a reference model Travis E. Gibson (gibson@mi.edu) Harvard Medical School Deparmen of Pahology, Brigham and Women s Hospial 55 h Conference

More information

Inventory Analysis and Management. Multi-Period Stochastic Models: Optimality of (s, S) Policy for K-Convex Objective Functions

Inventory Analysis and Management. Multi-Period Stochastic Models: Optimality of (s, S) Policy for K-Convex Objective Functions Muli-Period Sochasic Models: Opimali of (s, S) Polic for -Convex Objecive Funcions Consider a seing similar o he N-sage newsvendor problem excep ha now here is a fixed re-ordering cos (> 0) for each (re-)order.

More information

Languages That Are and Are Not Context-Free

Languages That Are and Are Not Context-Free Languages Tha re and re No Conex-Free Read K & S 3.5, 3.6, 3.7. Read Supplemenary Maerials: Conex-Free Languages and Pushdown uomaa: Closure Properies of Conex-Free Languages Read Supplemenary Maerials:

More information

Wrap up: Weighted, directed graph shortest path Minimum Spanning Tree. Feb 25, 2019 CSCI211 - Sprenkle

Wrap up: Weighted, directed graph shortest path Minimum Spanning Tree. Feb 25, 2019 CSCI211 - Sprenkle Objecive Wrap up: Weighed, direced graph hore pah Minimum Spanning Tree eb, 1 SI - Sprenkle 1 Review Wha are greedy algorihm? Wha i our emplae for olving hem? Review he la problem we were working on: Single-ource,

More information

Physics 235 Chapter 2. Chapter 2 Newtonian Mechanics Single Particle

Physics 235 Chapter 2. Chapter 2 Newtonian Mechanics Single Particle Chaper 2 Newonian Mechanics Single Paricle In his Chaper we will review wha Newon s laws of mechanics ell us abou he moion of a single paricle. Newon s laws are only valid in suiable reference frames,

More information

3, so θ = arccos

3, so θ = arccos Mahemaics 210 Professor Alan H Sein Monday, Ocober 1, 2007 SOLUTIONS This problem se is worh 50 poins 1 Find he angle beween he vecors (2, 7, 3) and (5, 2, 4) Soluion: Le θ be he angle (2, 7, 3) (5, 2,

More information

RL Lecture 7: Eligibility Traces. R. S. Sutton and A. G. Barto: Reinforcement Learning: An Introduction 1

RL Lecture 7: Eligibility Traces. R. S. Sutton and A. G. Barto: Reinforcement Learning: An Introduction 1 RL Lecure 7: Eligibiliy Traces R. S. Suon and A. G. Baro: Reinforcemen Learning: An Inroducion 1 N-sep TD Predicion Idea: Look farher ino he fuure when you do TD backup (1, 2, 3,, n seps) R. S. Suon and

More information

Two Popular Bayesian Estimators: Particle and Kalman Filters. McGill COMP 765 Sept 14 th, 2017

Two Popular Bayesian Estimators: Particle and Kalman Filters. McGill COMP 765 Sept 14 th, 2017 Two Popular Bayesian Esimaors: Paricle and Kalman Filers McGill COMP 765 Sep 14 h, 2017 1 1 1, dx x Bel x u x P x z P Recall: Bayes Filers,,,,,,, 1 1 1 1 u z u x P u z u x z P Bayes z = observaion u =

More information

1 Solutions to selected problems

1 Solutions to selected problems 1 Soluions o seleced problems 1. Le A B R n. Show ha in A in B bu in general bd A bd B. Soluion. Le x in A. Then here is ɛ > 0 such ha B ɛ (x) A B. This shows x in B. If A = [0, 1] and B = [0, 2], hen

More information

Suggested Practice Problems (set #2) for the Physics Placement Test

Suggested Practice Problems (set #2) for the Physics Placement Test Deparmen of Physics College of Ars and Sciences American Universiy of Sharjah (AUS) Fall 014 Suggesed Pracice Problems (se #) for he Physics Placemen Tes This documen conains 5 suggesed problems ha are

More information

6/3/2009. CS 244 Algorithm Design Instructor: t Artur Czumaj. Lecture 8 Network flows. Maximum Flow and Minimum Cut. Minimum Cut Problem.

6/3/2009. CS 244 Algorithm Design Instructor: t Artur Czumaj. Lecture 8 Network flows. Maximum Flow and Minimum Cut. Minimum Cut Problem. Maximum Flow and Minimum Cu CS lgorihm Deign Inrucor: rur Czumaj Lecure Nework Max and min cu. Two very rich algorihmic problem. Cornerone problem in combinaorial opimizaion. Beauiful mahemaical dualiy.

More information

Graphs III - Network Flow

Graphs III - Network Flow Graph III - Nework Flow Flow nework eup graph G=(V,E) edge capaciy w(u,v) 0 - if edge doe no exi, hen w(u,v)=0 pecial verice: ource verex ; ink verex - no edge ino and no edge ou of Aume every verex v

More information

Retrieval Models. Boolean and Vector Space Retrieval Models. Common Preprocessing Steps. Boolean Model. Boolean Retrieval Model

Retrieval Models. Boolean and Vector Space Retrieval Models. Common Preprocessing Steps. Boolean Model. Boolean Retrieval Model 1 Boolean and Vecor Space Rerieval Models Many slides in his secion are adaped from Prof. Joydeep Ghosh (UT ECE) who in urn adaped hem from Prof. Dik Lee (Univ. of Science and Tech, Hong Kong) Rerieval

More information

Delivering Better Time-of-Day Using Synchronous Ethernet and Yaakov (J) Stein, Alon Geva, Gabriel Zigelboim RAD Data Communications

Delivering Better Time-of-Day Using Synchronous Ethernet and Yaakov (J) Stein, Alon Geva, Gabriel Zigelboim RAD Data Communications Delivering Beer Time-of-Day Using Synchronous Eherne and 1588 Yaakov (J) Sein, Alon Geva, Gabriel Zigelboim RAD Daa Communicaions The problem I wan discuss he use of 1588 (or NTP for ha maer) for ime of

More information