Bayesian belief networks: learning and inference

Size: px
Start display at page:

Download "Bayesian belief networks: learning and inference"

Transcription

1 CS 1675 Introducton to chn Lrnng Lctur 16 ysn lf ntworks: lrnng nd nfrnc los Huskrcht 5329 Snnott Squr Dt: Dnsty stton D { D1 D2.. Dn} D x vctor of ttrut vlus Ojctv: try to stt th undrlyng tru prolty dstruton ovr vrls p usng xpls n D tru dstruton n spls p D D D.. D } { 1 2 n stt pˆ Stndrd d ssuptons: Spls r ndpndnt of ch othr co fro th s dntcl dstruton fxd p 1

2 odlng coplx dstrutons Quston: How to odl nd lrn coplx ultvrt dstrutons pˆ wth lrg nur of vrls? xpl: odlng of dss syptos rltons Dss: pnuon tnt syptos fndngs l tsts: vr Cough lnss WC wht lood clls count Chst pn tc. odl of th full jont dstruton: nuon vr Cough lnss WC Chst pn On prolty pr ssgnnt of vlus to vrls: nuon = vr = Cought= WC=Hgh Chst pn= ysn lf ntworks Ns ysn lf ntworks lt 80s gnnng of 90s Gv solutons to th spc cquston ottlncks rtl solutons for t coplxts Ky fturs: Rprsnt th full jont dstruton ovr th vrls or copctly wth sllr nur of prtrs. k dvntg of condtonl nd rgnl ndpndncs ong rndo vrls nd Y r ndpndnt Y Y nd Y r condtonlly ndpndnt gvn Z Y Z Z Y Z Y Z Z 2

3 ysn lf ntwork 1. Drctd cyclc grph Nods = rndo vrls urglry rthquk lr ry clls nd ohn clls Lnks = drct cusl dpndncs twn vrls. h chnc of lr ng s nfluncd y rthquk h chnc of ohn cllng s ffctd y th lr urglry rthquk lr ohnclls ryclls ysn lf ntwork 2. Locl condtonl dstrutons rltng vrls nd thr prnts urglry rthquk lr ohnclls ryclls 3

4 ysn lf ntwork urglry ohnclls lr rthquk ryclls ull jont dstruton n Ns ull jont dstruton s dfnd n trs of locl condtonl dstrutons otnd v th chn rul: n xpl: 1.. n ssu th followng ssgnnt of vlus to rndo vrls hn ts prolty s: p 4

5 5 ull jont dstruton n Ns Rwrt th full jont prolty usng th product rul: # of prtrs of th full jont: rtr coplxty prol In th N th full jont dstruton s dfnd s: Wht dd w sv? lr xpl: nry ru ls vrls urglry ohnclls lr rthquk ryclls n n p On prtr s for fr: # of prtrs of th N:?

6 ysn lf ntwork: prtrs count otl: 20 urglry ohnclls lr rthquk ryclls rtr coplxty prol In th N th full jont dstruton s dfnd s: n p 1.. n Wht dd w sv? lr xpl: 5 nry ru ls vrls # of prtrs of th full jont: urglry On prtr s for fr: # of prtrs of th N: ohnclls On prtr n vry condtonl s for fr:? lr rthquk ryclls 6

7 ysn lf ntwork: fr prtrs urglry otl fr prs: 10 ohnclls lr rthquk ryclls = = rtr coplxty prol In th N th full jont dstruton s dfnd s: n p 1.. n Wht dd w sv? lr xpl: 5 nry ru ls vrls # of prtrs of th full jont: urglry On prtr s for fr: # of prtrs of th N: ohnclls On prtr n vry condtonl s for fr: lr rthquk ryclls 7

8 Ns xpls In vrous rs: Intllgnt usr ntrfcs crosoft roulshootng dgnoss of tchncl dvc dcl dgnoss: thfndr CSC unn QR-D Collortv fltrng ltry pplctons Insurnc crdt pplctons Dgnoss of cr ngn Dgnos th ngn strt prol 8

9 Cr nsurnc xpl rdct cl costs dcl llty sd on pplcton dt ICU lr ntwork 9

10 CCS Coputr-sd tnt Cs Sulton syst CCS- dvlopd y rkr nd llr t Unvrsty of ttsurgh 422 nods nd 867 rcs QR-D dcl dgnoss n ntrnl dcn prtt ntwork of dss/fndngs rltons 10

11 11 Nïv ys odl spcl spl ysn lf ntwork usd s gnrtv clssfr odl odl of pxy = px y py Clss vrl y py ttruts r ndpndnt gvn y Lrnng: rtrz odls of py nd ll px j y= L stts of th prtrs Clss y 1 x 2 x n x.. 1 y x p y p d j j x Nïv ys odl spcl spl ysn lf ntwork usd s gnrtv clssfr odl odl of pxy = px y py Clssfcton: gvn x slct th clss Slct th clss wth th xu postror Clculton of postror s n xpl of N nfrnc Rr: w cn clcult th prolts fro th full jont Clss Y 1 2 n.. k u d j j d j j k u u y x p u y p y x p y p u y p u y p y p y p y p x x x

12 Lrnng of N Lrnng. Lrnng of prtrs of condtonl prolts Lrnng of th ntwork structur Vrls: Osrvl vlus prsnt n vry dt spl Hddn thy vlus r nvr osrvd n dt ssng vlus vlus sots prsnt sots not Nxt: Lrnng of th prtrs of N Vlus for ll vrls r osrvl stton of prtrs of N Id: dcopos th stton prol for th full jont ovr lrg nur of vrls to st of sllr stton prols corrspondng to locl prnt-vrl condtonls. xpl: ssu r nry wth ru ls vlus Lrnng of = 4 stton prols == == == == ssupton tht nls th dcoposton: prtrs of condtonl dstrutons r ndpndnt 12

13 stts of prtrs of N wo ssuptons tht prt th dcoposton: Spl ndpndnc D Θ Θ u1 rtr ndpndnc n N p Θ D p D q 1 j1 D u rtrs of ch condtonl on for vry ssgnnt of vlus to prnt vrls cn lrnd ndpndntly j # of nods # of prnts vlus Lrnng of N prtrs. xpl. xpl: nuon nuon?? HWCnu n???? lnss vr Cough Hgh WC lnnu vrnu Coughnu??? 13

14 Lrnng of N prtrs. xpl. Dt D dffrnt ptnt css: l v Cou HW nu lnss vr nuon Cough Hgh WC stts of prtrs of N uch lk ultpl con toss or roll of dc prols. sllr lrnng prol corrsponds to th lrnng of xctly on condtonl dstruton xpl: vr nuon rol: How to pck th dt to lrn? 14

15 Lrnng of N prtrs. xpl. Lrn: vr nuon Stp 1: Slct dt ponts wth nuon= l v Cou HW nu lnss vr nuon Cough Hgh WC Lrnng of N prtrs. xpl. Lrn: Stp 1: vr nuon Ignor th rst l v Cou HW nu lnss vr nuon Cough Hgh WC 15

16 Lrnng of N prtrs. xpl. Lrn: vr nuon Stp 2: Slct vlus of th rndo vrl dfnng th dstruton of vr l v Cou HW nu lnss vr nuon Cough Hgh WC Lrnng of N prtrs. xpl. Lrn: vr nuon Stp 2: Ignor th rst v lnss vr nuon Cough Hgh WC 16

17 Lrnng of N prtrs. xpl. Lrn: vr nuon Stp 3: Lrnng th L stt v lnss vr nuon Cough Hgh WC vr nuon Lrnng of N prtrs. ysn lrnng. Lrn: vr nuon Stp 3: Lrnng th ysn postror ssu th pror nuon vr nuon ~ t34 v lnss vr Cough Hgh WC ostror: vr nuon ~ t66 stts 6 1 vr nuon

18 stts of prtrs of N uch lk ultpl con toss or roll of dc prols. sllr lrnng prol corrsponds to th lrnng of xctly on condtonl dstruton xpl: vr nuon rol: How to pck th dt to lrn? nswr: 1. Slct dt ponts wth nuon= gnor th rst 2. ocus on slct only vlus of th rndo vrl dfnng th dstruton vr 3. Lrn th prtrs of th locl condtonls th s wy s w lrnd th prtrs of sd con or d rolstc nfrncs N odls copctly th full jont dstruton y tkng dvntg of xstng ndpndncs twn vrls Splfs th rprsntton nd lrnng of odl Cn usd for th dffrnt nfrnc tsks. 18

19 ys thor Condtonl/jont prolty rltons. ys thor swtchs condtonng vnts : Whn s t usful? Whn w r ntrstd n coputng th dgnostc qury fro th cusl prolty ffct cus cus cus ffct ffct Rson: It s oftn sr to ssss cusl prolty.g. rolty of pnuon cusng fvr vs. prolty of pnuon gvn fvr xpl: spl dgnostc nfrnc Dvc qupnt oprtng norlly or lfunctonng. Oprton of th dvc snsd ndrctly v snsor Snsor rdng s thr Hgh or Low N Dvc sttus Dvc sttus norl lfunctonng Snsor rdng Snsor rdng Dvc sttus Sttus\Snsor Hgh Low norl lfunc

20 xpl: spl dgnostc nfrnc Dgnostc nfrnc: coput th prolty of dvc oprtng norlly or lfunctonng gvn snsor rdng Dvc sttus Snsor rdng hgh Dvc sttus norl Snsor rdng hgh Dvc sttus lfunct Snsor rdng hgh Not w hv th oppost condtonl prolts: thy r uch sr to stt Soluton: pply ys thor to rvrs th condtonng vrls Dvc sttus Snsor rdng xpl: spl dgnostc nfrnc Dvc qupnt oprtng norlly or lfunctonng. Oprton of th dvc snsd ndrctly v snsor Snsor rdng s thr Hgh or Low Dvc sttus Dvc sttus norl lfunctonng Snsor rdng Dvc sttus Snsor rdng Sttus\Snsor Hgh Low norl lfunc Dvc sttus Snsor rdng hgh? 20

21 ys thor ssu vrl wth ultpl vlus ys thor cn rwrttn s: 1 2 k j j j k j 1 j j Usd n prctc whn w wnt to coput: for ll vlus of 1 2 k j j. xpl: spl dgnostc nfrnc Dvc sttus Dvc sttus norl lfunctonng Snsor rdng Snsor rdng Dvc sttus Sttus\Snsor Hgh Low norl lfunc Dvc sttus Snsor rdng hgh? Dvc sttus nor Snsor rdng hgh Snsor rdng hghdvc sttus nor Snsor rdng hgh 21

22 xpl: spl dgnostc nfrnc. Dvc sttus Dvc sttus norl lfunctonng Snsor rdng Snsor rdng Dvc sttus Sttus\Snsor Hgh Low norl lfunc Dvc sttus Snsor rdng hgh? Dvc sttus nor Snsor rdng hgh Snsor rdng hghdvc sttus nor Snsor rdng hgh Snsor rdng hgh Dvc sttus nor Dvc sttus nor Snsor rdng hgh xpl: spl dgnostc nfrnc. Dvc sttus Dvc sttus norl lfunctonng Snsor rdng Snsor rdng Dvc sttus Sttus\Snsor Hgh Low norl lfunc Dvc sttus Snsor rdng hgh? Dvc sttus nor Snsor rdng hgh Snsor rdng hgh Dvc sttus nor Dvc sttus nor Snsor rdng hgh Snsor rdng hgh Dvc sttus nor Dvc sttus nor Snsor rdng hgh Dvc sttus lf Dvc sttus lf 22

23 Infrnc n ysn ntworks N odls copctly th full jont dstruton y tkng dvntg of xstng ndpndncs twn vrls Splfs th rprsntton nd lrnng of odl ut w r ntrstd n solvng vrous nfrnc tsks: Dgnostc tsk. fro ffct to cus urglry ohnclls rdcton tsk. fro cus to ffct ohnclls urglry Othr prolstc qurs qurs on jont dstrutons. lr n quston: Cn w tk dvntg of ndpndncs to construct spcl lgorths nd spdng up th nfrnc? Infrnc n ysn ntwork d nws: xct nfrnc prol n Ns s N-hrd Coopr pproxt nfrnc s N-hrd Dgu Luy ut vry oftn w cn chv sgnfcnt provnts ssu our lr ntwork urglry rthquk lr ohnclls ryclls ssu w wnt to coput: 23

24 24 Infrnc n ysn ntworks Coputng: pproch 1. lnd pproch. Su out ll un-nstnttd vrls fro th full jont xprss th jont dstruton s product of condtonls Coputtonl cost: Nur of ddtons:? Nur of products:? Infrnc n ysn ntworks Coputng: pproch 1. lnd pproch. Su out ll un-nstnttd vrls fro th full jont xprss th jont dstruton s product of condtonls Coputtonl cost: Nur of ddtons: 15 Nur of products:?

25 25 Infrnc n ysn ntworks Coputng: pproch 1. lnd pproch. Su out ll un-nstnttd vrls fro th full jont xprss th jont dstruton s product of condtonls Coputtonl cost: Nur of ddtons: 15 Nur of products: 16*4=64 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons: 1+2*[1+1+2*1]=? Nur of products: 2*[2+2*1+2*1]=? ] [. ]] [ ][ [

26 26 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons:? 1 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons:? 2*1

27 27 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons:? 2*2*1 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons:? 2*2*1 2*1

28 28 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons:? 2*2*1 2*1 2*1 1 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons: 1+2*[1+1+2*1]=9 2*2*1 2*1 2*1 1

29 29 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of products:? 1 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of products:? 2*2 *2*1

30 30 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of products:? 2*2 *2*1 2*2*1 2*2 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of products: 2*[2+2*1+2*1]=16 2*2 *2*1 2*2*1 2*2

31 31 Infrnc n ysn ntworks pproch 2. Intrlv sus nd products Cons sus nd product n srt wy ultplctons y constnts cn tkn out of th su Coputtonl cost: Nur of ddtons: 1+2*[1+1+2*1]=9 Nur of products: 2*[2+2*1+2*1]=16. Vrl lnton Vrl lnton: Slr d ut ntrlv su nd products on vrl t th t durng nfrnc.g. Qury rqurs to lnt nd ths cn don n dffrnt ordr

32 32 Vrl lnton ssu ordr: to clcult Infrnc n ysn ntwork xct nfrnc lgorths: Vrl lnton Rcursv dcoposton Coopr Drwch Syolc nfrnc D roso lf propgton lgorth rl Clustrng nd jont tr pproch Lurtzn Spglhltr rc rvrsl Olstd Schchtr pproxt nfrnc lgorths: ont Crlo thods: orwrd splng Lklhood splng Vrtonl thods

Bayesian belief networks

Bayesian belief networks CS 1571 Introducton to I Lctur 20 ysn lf ntworks los Huskrcht los@cs.ptt.du 5329 Snnott Squr CS 1571 Intro to I. Huskrcht odlng uncrtnty wth prolts Dfnng th full jont dstruton ks t possl to rprsnt nd rson

More information

Bayesian belief networks: Inference

Bayesian belief networks: Inference C 740 Knowd rprntton ctur 0 n f ntwork: nfrnc o ukrcht o@c.ptt.du 539 nnott qur C 750 chn rnn n f ntwork. 1. Drctd ccc rph Nod rndo vr nk n nk ncod ndpndnc. urr rthquk r ohnc rc C 750 chn rnn n f ntwork.

More information

Bayesian belief networks

Bayesian belief networks CS 2750 oundtions of I Lctur 9 ysin lif ntworks ilos Huskrcht ilos@cs.pitt.du 5329 Snnott Squr. Huskrcht odling uncrtinty with proilitis Dfining th full joint distriution ks it possil to rprsnt nd rson

More information

Having a glimpse of some of the possibilities for solutions of linear systems, we move to methods of finding these solutions. The basic idea we shall

Having a glimpse of some of the possibilities for solutions of linear systems, we move to methods of finding these solutions. The basic idea we shall Hvn lps o so o t posslts or solutons o lnr systs, w ov to tos o nn ts solutons. T s w sll us s to try to sply t syst y lntn so o t vrls n so ts qutons. Tus, w rr to t to s lnton. T prry oprton nvolv s

More information

CIS587 - Artificial Intelligence. Uncertainty CIS587 - AI. KB for medical diagnosis. Example.

CIS587 - Artificial Intelligence. Uncertainty CIS587 - AI. KB for medical diagnosis. Example. CIS587 - rtfcl Intellgence Uncertnty K for medcl dgnoss. Exmple. We wnt to uld K system for the dgnoss of pneumon. rolem descrpton: Dsese: pneumon tent symptoms fndngs, l tests: Fever, Cough, leness, WC

More information

PH427/PH527: Periodic systems Spring Overview of the PH427 website (syllabus, assignments etc.) 2. Coupled oscillations.

PH427/PH527: Periodic systems Spring Overview of the PH427 website (syllabus, assignments etc.) 2. Coupled oscillations. Dy : Mondy 5 inuts. Ovrviw of th PH47 wsit (syllus, ssignnts tc.). Coupld oscilltions W gin with sss coupld y Hook's Lw springs nd find th possil longitudinl) otion of such syst. W ll xtnd this to finit

More information

Preview. Graph. Graph. Graph. Graph Representation. Graph Representation 12/3/2018. Graph Graph Representation Graph Search Algorithms

Preview. Graph. Graph. Graph. Graph Representation. Graph Representation 12/3/2018. Graph Graph Representation Graph Search Algorithms /3/0 Prvw Grph Grph Rprsntton Grph Srch Algorthms Brdth Frst Srch Corrctnss of BFS Dpth Frst Srch Mnmum Spnnng Tr Kruskl s lgorthm Grph Drctd grph (or dgrph) G = (V, E) V: St of vrt (nod) E: St of dgs

More information

Convergence Theorems for Two Iterative Methods. A stationary iterative method for solving the linear system: (1.1)

Convergence Theorems for Two Iterative Methods. A stationary iterative method for solving the linear system: (1.1) Conrgnc Thors for Two Itrt Mthods A sttonry trt thod for solng th lnr syst: Ax = b (.) ploys n trton trx B nd constnt ctor c so tht for gn strtng stt x of x for = 2... x Bx c + = +. (.2) For such n trton

More information

A Probabilistic Characterization of Simulation Model Uncertainties

A Probabilistic Characterization of Simulation Model Uncertainties A Proalstc Charactrzaton of Sulaton Modl Uncrtants Vctor Ontvros Mohaad Modarrs Cntr for Rsk and Rlalty Unvrsty of Maryland 1 Introducton Thr s uncrtanty n odl prdctons as wll as uncrtanty n xprnts Th

More information

Minimum Spanning Trees

Minimum Spanning Trees Mnmum Spnnng Trs Spnnng Tr A tr (.., connctd, cyclc grph) whch contns ll th vrtcs of th grph Mnmum Spnnng Tr Spnnng tr wth th mnmum sum of wghts 1 1 Spnnng forst If grph s not connctd, thn thr s spnnng

More information

INF5820/INF9820 LANGUAGE TECHNOLOGICAL APPLICATIONS. Jan Tore Lønning, Lecture 4, 14 Sep

INF5820/INF9820 LANGUAGE TECHNOLOGICAL APPLICATIONS. Jan Tore Lønning, Lecture 4, 14 Sep INF5820/INF9820 LANGUAGE TECHNOLOGICAL ALICATIONS Jn Tor Lønning Lctur 4 4 Sp. 206 tl@ii.uio.no Tody 2 Sttisticl chin trnsltion: Th noisy chnnl odl Word-bsd Trining IBM odl 3 SMT xpl 4 En kokk lgd n rtt

More information

SAMPLE CSc 340 EXAM QUESTIONS WITH SOLUTIONS: part 2

SAMPLE CSc 340 EXAM QUESTIONS WITH SOLUTIONS: part 2 AMPLE C EXAM UETION WITH OLUTION: prt. It n sown tt l / wr.7888l. I Φ nots orul or pprotng t vlu o tn t n sown tt t trunton rror o ts pproton s o t or or so onstnts ; tt s Not tt / L Φ L.. Φ.. /. /.. Φ..787.

More information

a b c cat CAT A B C Aa Bb Cc cat cat Lesson 1 (Part 1) Verbal lesson: Capital Letters Make The Same Sound Lesson 1 (Part 1) continued...

a b c cat CAT A B C Aa Bb Cc cat cat Lesson 1 (Part 1) Verbal lesson: Capital Letters Make The Same Sound Lesson 1 (Part 1) continued... Progrssiv Printing T.M. CPITLS g 4½+ Th sy, fun (n FR!) wy to tch cpitl lttrs. ook : C o - For Kinrgrtn or First Gr (not for pr-school). - Tchs tht cpitl lttrs mk th sm souns s th littl lttrs. - Tchs th

More information

5/1/2018. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees

5/1/2018. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees. Huffman Coding Trees /1/018 W usully no strns y ssnn -lnt os to ll rtrs n t lpt (or mpl, 8-t on n ASCII). Howvr, rnt rtrs our wt rnt rquns, w n sv mmory n ru trnsmttl tm y usn vrl-lnt non. T s to ssn sortr os to rtrs tt our

More information

Integration Continued. Integration by Parts Solving Definite Integrals: Area Under a Curve Improper Integrals

Integration Continued. Integration by Parts Solving Definite Integrals: Area Under a Curve Improper Integrals Intgrtion Continud Intgrtion y Prts Solving Dinit Intgrls: Ar Undr Curv Impropr Intgrls Intgrtion y Prts Prticulrly usul whn you r trying to tk th intgrl o som unction tht is th product o n lgric prssion

More information

On Hamiltonian Tetrahedralizations Of Convex Polyhedra

On Hamiltonian Tetrahedralizations Of Convex Polyhedra O Ht Ttrrzts O Cvx Pyr Frs C 1 Q-Hu D 2 C A W 3 1 Dprtt Cputr S T Uvrsty H K, H K, C. E: @s.u. 2 R & TV Trsss Ctr, Hu, C. E: q@163.t 3 Dprtt Cputr S, Mr Uvrsty Nwu St. J s, Nwu, C A1B 35. E: w@r.s.u. Astrt

More information

A Solution for multi-evaluator AHP

A Solution for multi-evaluator AHP ISAHP Honoll Hw Jly 8- A Solton for lt-vltor AHP Ms Shnohr Kch Osw Yo Hd Nhon Unvrsty Nhon Unvrsty Nhon Unvrsty Iz-cho Nrshno Iz-cho Nrshno Iz-cho Nrshno hb 7-87 Jpn hb 7-87 Jpn M7snoh@ct.nhon-.c.p 7oosw@ct.nhon-.c.p

More information

Constructing Free Energy Approximations and GBP Algorithms

Constructing Free Energy Approximations and GBP Algorithms 3710 Advnced Topcs n A ecture 15 Brnslv Kveton kveton@cs.ptt.edu 5802 ennott qure onstructng Free Energy Approxtons nd BP Algorths ontent Why? Belef propgton (BP) Fctor grphs egon-sed free energy pproxtons

More information

A general N-dimensional vector consists of N values. They can be arranged as a column or a row and can be real or complex.

A general N-dimensional vector consists of N values. They can be arranged as a column or a row and can be real or complex. Lnr lgr Vctors gnrl -dmnsonl ctor conssts of lus h cn rrngd s column or row nd cn rl or compl Rcll -dmnsonl ctor cn rprsnt poston, loct, or cclrton Lt & k,, unt ctors long,, & rspctl nd lt k h th componnts

More information

The Z transform techniques

The Z transform techniques h Z trnfor tchniqu h Z trnfor h th rol in dicrt yt tht th Lplc trnfor h in nlyi of continuou yt. h Z trnfor i th principl nlyticl tool for ingl-loop dicrt-ti yt. h Z trnfor h Z trnfor i to dicrt-ti yt

More information

Applications of trees

Applications of trees Trs Apptons o trs Orgnzton rts Attk trs to syst Anyss o tr ntworks Prsng xprssons Trs (rtrv o norton) Don-n strutur Mutstng Dstnton-s orwrng Trnsprnt swts Forwrng ts o prxs t routrs Struturs or nt pntton

More information

Fractions. Mathletics Instant Workbooks. Simplify. Copyright

Fractions. Mathletics Instant Workbooks. Simplify. Copyright Frctons Stunt Book - Srs H- Smplfy + Mthltcs Instnt Workbooks Copyrht Frctons Stunt Book - Srs H Contnts Topcs Topc - Equvlnt frctons Topc - Smplfyn frctons Topc - Propr frctons, mpropr frctons n mx numbrs

More information

ADORO TE DEVOTE (Godhead Here in Hiding) te, stus bat mas, la te. in so non mor Je nunc. la in. tis. ne, su a. tum. tas: tur: tas: or: ni, ne, o:

ADORO TE DEVOTE (Godhead Here in Hiding) te, stus bat mas, la te. in so non mor Je nunc. la in. tis. ne, su a. tum. tas: tur: tas: or: ni, ne, o: R TE EVTE (dhd H Hdg) L / Mld Kbrd gú s v l m sl c m qu gs v nns V n P P rs l mul m d lud 7 súb Fí cón ví f f dó, cru gs,, j l f c r s m l qum t pr qud ct, us: ns,,,, cs, cut r l sns m / m fí hó sn sí

More information

Section 3: Antiderivatives of Formulas

Section 3: Antiderivatives of Formulas Chptr Th Intgrl Appli Clculus 96 Sction : Antirivtivs of Formuls Now w cn put th is of rs n ntirivtivs togthr to gt wy of vluting finit intgrls tht is ct n oftn sy. To vlut finit intgrl f(t) t, w cn fin

More information

Computer Graphics. Viewing & Projections

Computer Graphics. Viewing & Projections Vw & Ovrvw rr : rss r t -vw trsrt: st st, rr w.r.t. r rqurs r rr (rt syst) rt: 2 trsrt st, rt trsrt t 2D rqurs t r y rt rts ss Rr P usuy st try trsrt t wr rts t rs t surs trsrt t r rts u rt w.r.t. vw vu

More information

Inner Product Spaces INNER PRODUCTS

Inner Product Spaces INNER PRODUCTS MA4Hcdoc Ir Product Spcs INNER PRODCS Dto A r product o vctor spc V s ucto tht ssgs ubr spc V such wy tht th ollowg xos holds: P : w s rl ubr P : P : P 4 : P 5 : v, w = w, v v + w, u = u + w, u rv, w =

More information

Minimum Spanning Trees

Minimum Spanning Trees Minimum Spnning Trs Minimum Spnning Trs Problm A town hs st of houss nd st of rods A rod conncts nd only houss A rod conncting houss u nd v hs rpir cost w(u, v) Gol: Rpir nough (nd no mor) rods such tht:

More information

The Mathematics of Harmonic Oscillators

The Mathematics of Harmonic Oscillators Th Mhcs of Hronc Oscllors Spl Hronc Moon In h cs of on-nsonl spl hronc oon (SHM nvolvng sprng wh sprng consn n wh no frcon, you rv h quon of oon usng Nwon's scon lw: con wh gvs: 0 Ths s sos wrn usng h

More information

CIVL 8/ D Boundary Value Problems - Rectangular Elements 1/7

CIVL 8/ D Boundary Value Problems - Rectangular Elements 1/7 CIVL / -D Boundr Vlu Prolms - Rctngulr Elmnts / RECANGULAR ELEMENS - In som pplictions, it m mor dsirl to us n lmntl rprsnttion of th domin tht hs four sids, ithr rctngulr or qudriltrl in shp. Considr

More information

Divided. diamonds. Mimic the look of facets in a bracelet that s deceptively deep RIGHT-ANGLE WEAVE. designed by Peggy Brinkman Matteliano

Divided. diamonds. Mimic the look of facets in a bracelet that s deceptively deep RIGHT-ANGLE WEAVE. designed by Peggy Brinkman Matteliano RIGHT-ANGLE WEAVE Dv mons Mm t look o ts n rlt tt s ptvly p sn y Py Brnkmn Mttlno Dv your mons nto trnls o two or our olors. FCT-SCON0216_BNB66 2012 Klm Pulsn Co. Ts mtrl my not rprou n ny orm wtout prmsson

More information

L.3922 M.C. L.3922 M.C. L.2996 M.C. L.3909 M.C. L.5632 M.C. L M.C. L.5632 M.C. L M.C. DRIVE STAR NORTH STAR NORTH NORTH DRIVE

L.3922 M.C. L.3922 M.C. L.2996 M.C. L.3909 M.C. L.5632 M.C. L M.C. L.5632 M.C. L M.C. DRIVE STAR NORTH STAR NORTH NORTH DRIVE N URY T NORTON PROV N RRONOUS NORTON NVRTNTY PROV. SPY S NY TY OR UT T TY RY OS NOT URNT T S TT T NORTON PROV S ORRT, NSR S POSS, VRY ORT S N ON N T S T TY RY. TS NORTON S N OP RO RORS RT SU "" YW No.

More information

GUC (Dr. Hany Hammad) 9/28/2016

GUC (Dr. Hany Hammad) 9/28/2016 U (r. Hny Hd) 9/8/06 ctur # 3 ignl flow grphs (cont.): ignl-flow grph rprsnttion of : ssiv sgl-port dvic. owr g qutions rnsducr powr g. Oprtg powr g. vill powr g. ppliction to Ntwork nlyzr lirtion. Nois

More information

MM1. Introduction to State-Space Method

MM1. Introduction to State-Space Method MM Itroductio to Stt-Spc Mthod Wht tt-pc thod? How to gt th tt-pc dcriptio? 3 Proprty Alyi Bd o SS Modl Rdig Mtril: FC: p469-49 C: p- /4/8 Modr Cotrol Wht th SttS tt-spc Mthod? I th tt-pc thod th dyic

More information

Weighted Graphs. Weighted graphs may be either directed or undirected.

Weighted Graphs. Weighted graphs may be either directed or undirected. 1 In mny ppltons, o rp s n ssot numrl vlu, ll wt. Usully, t wts r nonntv ntrs. Wt rps my tr rt or unrt. T wt o n s otn rrr to s t "ost" o t. In ppltons, t wt my msur o t lnt o rout, t pty o ln, t nry rqur

More information

An Ising model on 2-D image

An Ising model on 2-D image School o Coputer Scence Approte Inerence: Loopy Bele Propgton nd vrnts Prolstc Grphcl Models 0-708 Lecture 4, ov 7, 007 Receptor A Knse C Gene G Receptor B Knse D Knse E 3 4 5 TF F 6 Gene H 7 8 Hetunndn

More information

Extension Formulas of Lauricella s Functions by Applications of Dixon s Summation Theorem

Extension Formulas of Lauricella s Functions by Applications of Dixon s Summation Theorem Avll t http:pvu.u Appl. Appl. Mth. ISSN: 9-9466 Vol. 0 Issu Dr 05 pp. 007-08 Appltos Appl Mthts: A Itrtol Jourl AAM Etso oruls of Lurll s utos Appltos of Do s Suto Thor Ah Al Atsh Dprtt of Mthts A Uvrst

More information

minimize c'x subject to subject to subject to

minimize c'x subject to subject to subject to z ' sut to ' M ' M N uostrd N z ' sut to ' z ' sut to ' sl vrls vtor of : vrls surplus vtor of : uostrd s s s s s s z sut to whr : ut ost of :out of : out of ( ' gr of h food ( utrt : rqurt for h utrt

More information

(A) the function is an eigenfunction with eigenvalue Physical Chemistry (I) First Quiz

(A) the function is an eigenfunction with eigenvalue Physical Chemistry (I) First Quiz 96- Physcl Chmstry (I) Frst Quz lctron rst mss m 9.9 - klogrm, Plnck constnt h 6.66-4 oul scon Sp of lght c. 8 m/s, lctron volt V.6-9 oul. Th functon F() C[cos()+sn()] s n gnfuncton of /. Th gnvlu s (A)

More information

CSE 373: More on graphs; DFS and BFS. Michael Lee Wednesday, Feb 14, 2018

CSE 373: More on graphs; DFS and BFS. Michael Lee Wednesday, Feb 14, 2018 CSE 373: Mor on grphs; DFS n BFS Mihl L Wnsy, F 14, 2018 1 Wrmup Wrmup: Disuss with your nighor: Rmin your nighor: wht is simpl grph? Suppos w hv simpl, irt grph with x nos. Wht is th mximum numr of gs

More information

Wave Phenomena Physics 15c

Wave Phenomena Physics 15c Wv hnon hyscs 5c cur 4 Coupl Oscllors! H& con 4. Wh W D s T " u forc oscllon " olv h quon of oon wh frcon n foun h sy-s soluon " Oscllon bcos lr nr h rsonnc frquncy " hs chns fro 0 π/ π s h frquncy ncrss

More information

4.1 Interval Scheduling. Chapter 4. Greedy Algorithms. Interval Scheduling: Greedy Algorithms. Interval Scheduling. Interval scheduling.

4.1 Interval Scheduling. Chapter 4. Greedy Algorithms. Interval Scheduling: Greedy Algorithms. Interval Scheduling. Interval scheduling. Cptr 4 4 Intrvl Suln Gry Alortms Sls y Kvn Wyn Copyrt 005 Prson-Ason Wsly All rts rsrv Intrvl Suln Intrvl Suln: Gry Alortms Intrvl suln! Jo strts t s n nss t! Two os omptl ty on't ovrlp! Gol: n mxmum sust

More information

Last time: introduced our first computational model the DFA.

Last time: introduced our first computational model the DFA. Lctur 7 Homwork #7: 2.2.1, 2.2.2, 2.2.3 (hnd in c nd d), Misc: Givn: M, NFA Prov: (q,xy) * (p,y) iff (q,x) * (p,) (follow proof don in clss tody) Lst tim: introducd our first computtionl modl th DFA. Tody

More information

INTEGRALS. Chapter 7. d dx. 7.1 Overview Let d dx F (x) = f (x). Then, we write f ( x)

INTEGRALS. Chapter 7. d dx. 7.1 Overview Let d dx F (x) = f (x). Then, we write f ( x) Chptr 7 INTEGRALS 7. Ovrviw 7.. Lt d d F () f (). Thn, w writ f ( ) d F () + C. Ths intgrls r clld indfinit intgrls or gnrl intgrls, C is clld constnt of intgrtion. All ths intgrls diffr y constnt. 7..

More information

Outlier-tolerant parameter estimation

Outlier-tolerant parameter estimation Outlr-tolrant paramtr stmaton Baysan thods n physcs statstcs machn larnng and sgnal procssng (SS 003 Frdrch Fraundorfr fraunfr@cg.tu-graz.ac.at Computr Graphcs and Vson Graz Unvrsty of Tchnology Outln

More information

STATISTICAL MECHANICS OF THE INVERSE ISING MODEL

STATISTICAL MECHANICS OF THE INVERSE ISING MODEL STATISTICAL MECHANICS OF THE INVESE ISING MODEL Muro Cro Supervsors: rof. Mchele Cselle rof. ccrdo Zecchn uly 2009 INTODUCTION SUMMAY OF THE ESENTATION Defnton of the drect nd nverse prole Approton ethods

More information

The Theory of Small Reflections

The Theory of Small Reflections Jim Stils Th Univ. of Knss Dt. of EECS 4//9 Th Thory of Smll Rflctions /9 Th Thory of Smll Rflctions Rcll tht w nlyzd qurtr-wv trnsformr usg th multil rflction viw ot. V ( z) = + β ( z + ) V ( z) = = R

More information

Fundamentals of Continuum Mechanics. Seoul National University Graphics & Media Lab

Fundamentals of Continuum Mechanics. Seoul National University Graphics & Media Lab Fndmntls of Contnm Mchncs Sol Ntonl Unvrsty Grphcs & Md Lb Th Rodmp of Contnm Mchncs Strss Trnsformton Strn Trnsformton Strss Tnsor Strn T + T ++ T Strss-Strn Rltonshp Strn Enrgy FEM Formlton Lt s Stdy

More information

10/7/14. Mixture Models. Comp 135 Introduction to Machine Learning and Data Mining. Maximum likelihood estimation. Mixture of Normals in 1D

10/7/14. Mixture Models. Comp 135 Introduction to Machine Learning and Data Mining. Maximum likelihood estimation. Mixture of Normals in 1D Comp 35 Introducton to Machn Larnng and Data Mnng Fall 204 rofssor: Ron Khardon Mxtur Modls Motvatd by soft k-mans w dvlopd a gnratv modl for clustrng. Assum thr ar k clustrs Clustrs ar not rqurd to hav

More information

Multi-Section Coupled Line Couplers

Multi-Section Coupled Line Couplers /0/009 MultiSction Coupld Lin Couplrs /8 Multi-Sction Coupld Lin Couplrs W cn dd multipl coupld lins in sris to incrs couplr ndwidth. Figur 7.5 (p. 6) An N-sction coupld lin l W typiclly dsign th couplr

More information

Section 5.1/5.2: Areas and Distances the Definite Integral

Section 5.1/5.2: Areas and Distances the Definite Integral Scto./.: Ars d Dstcs th Dt Itgrl Sgm Notto Prctc HW rom Stwrt Ttook ot to hd p. #,, 9 p. 6 #,, 9- odd, - odd Th sum o trms,,, s wrtt s, whr th d o summto Empl : Fd th sum. Soluto: Th Dt Itgrl Suppos w

More information

CS September 2018

CS September 2018 Loil los Distriut Systms 06. Loil los Assin squn numrs to msss All ooprtin prosss n r on orr o vnts vs. physil los: rport tim o y Assum no ntrl tim sour Eh systm mintins its own lol lo No totl orrin o

More information

THE INFLUENCE OF HINDCAST MODELING UNCERTAINTY ON THE PREDICTION OF HIGH RETURN PERIOD WAVE CONDITIONS

THE INFLUENCE OF HINDCAST MODELING UNCERTAINTY ON THE PREDICTION OF HIGH RETURN PERIOD WAVE CONDITIONS Prodngs of OMAE4 3rd Intrntonl Confrn on Offshor Mhns nd Art Engnrng Jun -5, 4, Vnouvr, Brtsh Colub, Cnd OM AE4-56 THE INFLUENCE OF HINDCAST MODELING UNCERTAINTY ON THE PREDICTION OF HIGH RETURN PERIOD

More information

COMP108 Algorithmic Foundations

COMP108 Algorithmic Foundations Grdy mthods Prudn Wong http://www.s.liv..uk/~pwong/thing/omp108/01617 Coin Chng Prolm Suppos w hv 3 typs of oins 10p 0p 50p Minimum numr of oins to mk 0.8, 1.0, 1.? Grdy mthod Lrning outoms Undrstnd wht

More information

Review - Probabilistic Classification

Review - Probabilistic Classification Mmoral Unvrsty of wfoundland Pattrn Rcognton Lctur 8 May 5, 6 http://www.ngr.mun.ca/~charlsr Offc Hours: Tusdays Thursdays 8:3-9:3 PM E- (untl furthr notc) Gvn lablld sampls { ɛc,,,..., } {. Estmat Rvw

More information

Rate of Molecular Exchange Through the Membranes of Ionic Liquid Filled. Polymersomes Dispersed in Water

Rate of Molecular Exchange Through the Membranes of Ionic Liquid Filled. Polymersomes Dispersed in Water Supportng Informton for: Rt of Molculr Exchng hrough th Mmrns of Ionc Lqud Flld olymrsoms Dsprsd n Wtr Soonyong So nd mothy. Lodg *,, Dprtmnt of Chmcl Engnrng & Mtrls Scnc nd Dprtmnt of Chmstry, Unvrsty

More information

FL/VAL ~RA1::1. Professor INTERVI of. Professor It Fr recru. sor Social,, first of all, was. Sys SDC? Yes, as a. was a. assumee.

FL/VAL ~RA1::1. Professor INTERVI of. Professor It Fr recru. sor Social,, first of all, was. Sys SDC? Yes, as a. was a. assumee. B Pror NTERV FL/VAL ~RA1::1 1 21,, 1989 i n or Socil,, fir ll, Pror Fr rcru Sy Ar you lir SDC? Y, om um SM: corr n 'd m vry ummr yr. Now, y n y, f pr my ry for ummr my 1 yr Un So vr ummr cour d rr o l

More information

INF5820 MT 26 OCT 2012

INF5820 MT 26 OCT 2012 INF582 MT 26 OCT 22 H22 Jn Tor Lønnng l@.uo.no Tody Ssl hn rnslon: Th nosy hnnl odl Word-bsd IBM odl Trnng SMT xpl En o lgd n r d bygg..9 h.6 d.3.9 rgh.9 wh.4 buldng.45 oo.3 rd.25 srgh.7 by.3 onsruon.33

More information

The University of Sydney MATH 2009

The University of Sydney MATH 2009 T Unvrsty o Syny MATH 2009 APH THEOY Tutorl 7 Solutons 2004 1. Lt t sonnt plnr rp sown. Drw ts ul, n t ul o t ul ( ). Sow tt s sonnt plnr rp, tn s onnt. Du tt ( ) s not somorp to. ( ) A onnt rp s on n

More information

Ch 1.2: Solutions of Some Differential Equations

Ch 1.2: Solutions of Some Differential Equations Ch 1.2: Solutions of Som Diffrntil Equtions Rcll th fr fll nd owl/mic diffrntil qutions: v 9.8.2v, p.5 p 45 Ths qutions hv th gnrl form y' = y - b W cn us mthods of clculus to solv diffrntil qutions of

More information

INF5820/INF9820 LANGUAGE TECHNOLOGICAL APPLICATIONS. Jan Tore Lønning, Lecture 4, 10 Sep.

INF5820/INF9820 LANGUAGE TECHNOLOGICAL APPLICATIONS. Jan Tore Lønning, Lecture 4, 10 Sep. INF5820/INF9820 LANGUAGE TECHNOLOGICAL ALICATIONS Jn Tor Lønning Lctur 4 0 Sp. tl@ii.uio.no Tody 2 Sttisticl chin trnsltion: Th noisy chnnl odl Word-bsd Trining IBM odl 3 SMT xpl 4 En kokk lgd n rtt d

More information

Let's revisit conditional probability, where the event M is expressed in terms of the random variable. P Ax x x = =

Let's revisit conditional probability, where the event M is expressed in terms of the random variable. P Ax x x = = L's rvs codol rol whr h v M s rssd rs o h rdo vrl. L { M } rrr v such h { M } Assu. { } { A M} { A { } } M < { } { } A u { } { } { A} { A} ( A) ( A) { A} A A { A } hs llows us o cosdr h cs wh M { } [ (

More information

Lecture 11 Waves in Periodic Potentials Today: Questions you should be able to address after today s lecture:

Lecture 11 Waves in Periodic Potentials Today: Questions you should be able to address after today s lecture: Lctur 11 Wvs in Priodic Potntils Tody: 1. Invrs lttic dfinition in 1D.. rphicl rprsnttion of priodic nd -priodic functions using th -xis nd invrs lttic vctors. 3. Sris solutions to th priodic potntil Hmiltonin

More information

SAMPLE LITANY OF THE SAINTS E/G. Dadd9/F. Aadd9. cy. Christ, have. Lord, have mer cy. Christ, have A/E. Dadd9. Aadd9/C Bm E. 1. Ma ry and. mer cy.

SAMPLE LITANY OF THE SAINTS E/G. Dadd9/F. Aadd9. cy. Christ, have. Lord, have mer cy. Christ, have A/E. Dadd9. Aadd9/C Bm E. 1. Ma ry and. mer cy. LTNY OF TH SNTS Cntrs Gnt flwng ( = c. 100) /G Ddd9/F ll Kybrd / hv Ddd9 hv hv Txt 1973, CL. ll rghts rsrvd. Usd wth prmssn. Musc: D. Bckr, b. 1953, 1987, D. Bckr. Publshd by OCP. ll rghts rsrvd. SMPL

More information

Theorem 1. An undirected graph is a tree if and only if there is a unique simple path between any two of its vertices.

Theorem 1. An undirected graph is a tree if and only if there is a unique simple path between any two of its vertices. Cptr 11: Trs 11.1 - Introuton to Trs Dnton 1 (Tr). A tr s onnt unrt rp wt no sp ruts. Tor 1. An unrt rp s tr n ony tr s unqu sp pt twn ny two o ts vrts. Dnton 2. A root tr s tr n w on vrtx s n snt s t

More information

A L A BA M A L A W R E V IE W

A L A BA M A L A W R E V IE W A L A BA M A L A W R E V IE W Volume 52 Fall 2000 Number 1 B E F O R E D I S A B I L I T Y C I V I L R I G HT S : C I V I L W A R P E N S I O N S A N D TH E P O L I T I C S O F D I S A B I L I T Y I N

More information

Elliptical motion, gravity, etc

Elliptical motion, gravity, etc FW Physics 130 G:\130 lctur\ch 13 Elliticl motion.docx g 1 of 7 11/3/010; 6:40 PM; Lst rintd 11/3/010 6:40:00 PM Fig. 1 Elliticl motion, grvity, tc minor xis mjor xis F 1 =A F =B C - D, mjor nd minor xs

More information

Basic Electrical Engineering for Welding [ ] --- Introduction ---

Basic Electrical Engineering for Welding [ ] --- Introduction --- Basc Elctrcal Engnrng for Wldng [] --- Introducton --- akayosh OHJI Profssor Ertus, Osaka Unrsty Dr. of Engnrng VIUAL WELD CO.,LD t-ohj@alc.co.jp OK 15 Ex. Basc A.C. crcut h fgurs n A-group show thr typcal

More information

QUESTIONS BEGIN HERE!

QUESTIONS BEGIN HERE! Points miss: Stunt's Nm: Totl sor: /100 points Est Tnnss Stt Univrsity Dprtmnt o Computr n Inormtion Sins CSCI 2710 (Trno) Disrt Struturs TEST or Sprin Smstr, 2005 R this or strtin! This tst is los ook

More information

COMP 465: Data Mining More on PageRank

COMP 465: Data Mining More on PageRank COMP 465: Dt Mnng Moe on PgeRnk Sldes Adpted Fo: www.ds.og (Mnng Mssve Dtsets) Powe Iteton: Set = 1/ 1: = 2: = Goto 1 Exple: d 1/3 1/3 5/12 9/24 6/15 = 1/3 3/6 1/3 11/24 6/15 1/3 1/6 3/12 1/6 3/15 Iteton

More information

EE1000 Project 4 Digital Volt Meter

EE1000 Project 4 Digital Volt Meter Ovrviw EE1000 Projt 4 Diitl Volt Mtr In this projt, w mk vi tht n msur volts in th rn o 0 to 4 Volts with on iit o ury. Th input is n nlo volt n th output is sinl 7-smnt iit tht tlls us wht tht input s

More information

GEORGE F. JOWETT. HOLDER -of NUMEROUS DIPLOMAS and GOLD. MEDALS for ACTUAL MERIT

GEORGE F. JOWETT. HOLDER -of NUMEROUS DIPLOMAS and GOLD. MEDALS for ACTUAL MERIT GEORGE F OWE ANADAS SRONGES AHLEE HOLDER of NUMEROUS DPLOMAS nd GOLD MEDALS for AUAL MER AUHOR LEURER AND REOGNZED AUHORY ON PHYSAL EDUAON NKERMAN ONARO ANADA P : 6 23 D:::r P ul lv:; j"3: t your ltr t:

More information

Bayesian Networks: Approximate Inference

Bayesian Networks: Approximate Inference pproches to inference yesin Networks: pproximte Inference xct inference Vrillimintion Join tree lgorithm pproximte inference Simplify the structure of the network to mkxct inferencfficient (vritionl methods,

More information

ECE COMBINATIONAL BUILDING BLOCKS - INVEST 13 DECODERS AND ENCODERS

ECE COMBINATIONAL BUILDING BLOCKS - INVEST 13 DECODERS AND ENCODERS C 24 - COMBINATIONAL BUILDING BLOCKS - INVST 3 DCODS AND NCODS FALL 23 AP FLZ To o "wll" on this invstition you must not only t th riht nswrs ut must lso o nt, omplt n onis writups tht mk ovious wht h

More information

Soft k-means Clustering. Comp 135 Machine Learning Computer Science Tufts University. Mixture Models. Mixture of Normals in 1D

Soft k-means Clustering. Comp 135 Machine Learning Computer Science Tufts University. Mixture Models. Mixture of Normals in 1D Comp 35 Machn Larnng Computr Scnc Tufts Unvrsty Fall 207 Ron Khardon Th EM Algorthm Mxtur Modls Sm-Suprvsd Larnng Soft k-mans Clustrng ck k clustr cntrs : Assocat xampls wth cntrs p,j ~~ smlarty b/w cntr

More information

Spanning Tree. Preview. Minimum Spanning Tree. Minimum Spanning Tree. Minimum Spanning Tree. Minimum Spanning Tree 10/17/2017.

Spanning Tree. Preview. Minimum Spanning Tree. Minimum Spanning Tree. Minimum Spanning Tree. Minimum Spanning Tree 10/17/2017. 0//0 Prvw Spnnng Tr Spnnng Tr Mnmum Spnnng Tr Kruskl s Algorthm Prm s Algorthm Corrctnss of Kruskl s Algorthm A spnnng tr T of connctd, undrctd grph G s tr composd of ll th vrtcs nd som (or prhps ll) of

More information

TOPIC 5: INTEGRATION

TOPIC 5: INTEGRATION TOPIC 5: INTEGRATION. Th indfinit intgrl In mny rspcts, th oprtion of intgrtion tht w r studying hr is th invrs oprtion of drivtion. Dfinition.. Th function F is n ntidrivtiv (or primitiv) of th function

More information

Linear Algebra Existence of the determinant. Expansion according to a row.

Linear Algebra Existence of the determinant. Expansion according to a row. Lir Algbr 2270 1 Existc of th dtrmit. Expsio ccordig to row. W dfi th dtrmit for 1 1 mtrics s dt([]) = (1) It is sy chck tht it stisfis D1)-D3). For y othr w dfi th dtrmit s follows. Assumig th dtrmit

More information

b.) v d =? Example 2 l = 50 m, D = 1.0 mm, E = 6 V, " = 1.72 #10 $8 % & m, and r = 0.5 % a.) R =? c.) V ab =? a.) R eq =?

b.) v d =? Example 2 l = 50 m, D = 1.0 mm, E = 6 V,  = 1.72 #10 $8 % & m, and r = 0.5 % a.) R =? c.) V ab =? a.) R eq =? xmpl : An 8-gug oppr wr hs nomnl mtr o. mm. Ths wr rrs onstnt urrnt o.67 A to W lmp. Th nsty o r ltrons s 8.5 x 8 ltrons pr u mtr. Fn th mgntu o. th urrnt nsty. th rt vloty xmpl D. mm,.67 A, n N 8.5" 8

More information

1. Accident preve. 3. First aid kit ess 4. ABCs of life do. 6. Practice a Build a pasta sk

1. Accident preve. 3. First aid kit ess 4. ABCs of life do. 6. Practice a Build a pasta sk Y M D B D K P S V P U D hi p r ub g rup ck l yu cn 7 r, f r i y un civi i u ir r ub c fr ll y u n rgncy i un pg 3-9 bg i pr hich. ff c cn b ll p i f h grup r b n n c rk ivii ru gh g r! i pck? i i rup civ

More information

Lecture 20: Minimum Spanning Trees (CLRS 23)

Lecture 20: Minimum Spanning Trees (CLRS 23) Ltur 0: Mnmum Spnnn Trs (CLRS 3) Jun, 00 Grps Lst tm w n (wt) rps (unrt/rt) n ntrou s rp voulry (vrtx,, r, pt, onnt omponnts,... ) W lso suss jny lst n jny mtrx rprsntton W wll us jny lst rprsntton unlss

More information

Decimals DECIMALS.

Decimals DECIMALS. Dimls DECIMALS www.mthltis.o.uk ow os it work? Solutions Dimls P qustions Pl vlu o imls 0 000 00 000 0 000 00 0 000 00 0 000 00 0 000 tnths or 0 thousnths or 000 hunrths or 00 hunrths or 00 0 tn thousnths

More information

FSA. CmSc 365 Theory of Computation. Finite State Automata and Regular Expressions (Chapter 2, Section 2.3) ALPHABET operations: U, concatenation, *

FSA. CmSc 365 Theory of Computation. Finite State Automata and Regular Expressions (Chapter 2, Section 2.3) ALPHABET operations: U, concatenation, * CmSc 365 Thory of Computtion Finit Stt Automt nd Rgulr Exprssions (Chptr 2, Sction 2.3) ALPHABET oprtions: U, conctntion, * otin otin Strings Form Rgulr xprssions dscri Closd undr U, conctntion nd * (if

More information

Machine learning: Density estimation

Machine learning: Density estimation CS 70 Foundatons of AI Lecture 3 Machne learnng: ensty estmaton Mlos Hauskrecht mlos@cs.ptt.edu 539 Sennott Square ata: ensty estmaton {.. n} x a vector of attrbute values Objectve: estmate the model of

More information

n r t d n :4 T P bl D n, l d t z d th tr t. r pd l

n r t d n :4 T P bl D n, l d t z d   th tr t. r pd l n r t d n 20 20 :4 T P bl D n, l d t z d http:.h th tr t. r pd l 2 0 x pt n f t v t, f f d, b th n nd th P r n h h, th r h v n t b n p d f r nt r. Th t v v d pr n, h v r, p n th pl v t r, d b p t r b R

More information

, between the vertical lines x a and x b. Given a demand curve, having price as a function of quantity, p f (x) at height k is the curve f ( x,

, between the vertical lines x a and x b. Given a demand curve, having price as a function of quantity, p f (x) at height k is the curve f ( x, Clculus for Businss nd Socil Scincs - Prof D Yun Finl Em Rviw vrsion 5/9/7 Chck wbsit for ny postd typos nd updts Pls rport ny typos This rviw sht contins summris of nw topics only (This rviw sht dos hv

More information

Planar convex hulls (I)

Planar convex hulls (I) Covx Hu Covxty Gv st P o ots 2D, tr ovx u s t sst ovx oyo tt ots ots o P A oyo P s ovx or y, P, t st s try P. Pr ovx us (I) Coutto Gotry [s 3250] Lur To Bowo Co ovx o-ovx 1 2 3 Covx Hu Covx Hu Covx Hu

More information

Grand Canonical Ensemble

Grand Canonical Ensemble Th nsmbl of systms mmrsd n a partcl-hat rsrvor at constant tmpratur T, prssur P, and chmcal potntal. Consdr an nsmbl of M dntcal systms (M =,, 3,...M).. Thy ar mutually sharng th total numbr of partcls

More information

Chapter 16. 1) is a particular point on the graph of the function. 1. y, where x y 1

Chapter 16. 1) is a particular point on the graph of the function. 1. y, where x y 1 Prctic qustions W now tht th prmtr p is dirctl rltd to th mplitud; thrfor, w cn find tht p. cos d [ sin ] sin sin Not: Evn though ou might not now how to find th prmtr in prt, it is lws dvisl to procd

More information

Eastern Progress - 3 Mar 1923

Eastern Progress - 3 Mar 1923 922-927 Kk U Y 923-3 923 Kk U //k/ 922-27/7 N VOLU WO X-COON O DUCON OG COND DON COUNY W WOOD LCD DO O NNUL NOC - N CL WNGON DY D Cx W Oz N WN GN O U N N C U D Y C 3 923 CUC OCL W NOD VN W C 9 NO OU UDN

More information

In which direction do compass needles always align? Why?

In which direction do compass needles always align? Why? AQA Trloy Unt 6.7 Mntsm n Eltromntsm - Hr 1 Complt t p ll: Mnt or s typ o or n t s stronst t t o t mnt. Tr r two typs o mnt pol: n. Wrt wt woul ppn twn t pols n o t mnt ntrtons low: Drw t mnt l lns on

More information

Graphs. CSC 1300 Discrete Structures Villanova University. Villanova CSC Dr Papalaskari

Graphs. CSC 1300 Discrete Structures Villanova University. Villanova CSC Dr Papalaskari Grphs CSC 1300 Disrt Struturs Villnov Univrsity Grphs Grphs r isrt struturs onsis?ng of vr?s n gs tht onnt ths vr?s. Grphs n us to mol: omputr systms/ntworks mthm?l rl?ons logi iruit lyout jos/prosss f

More information

ELEG 413 Lecture #6. Mark Mirotznik, Ph.D. Professor The University of Delaware

ELEG 413 Lecture #6. Mark Mirotznik, Ph.D. Professor The University of Delaware LG 43 Lctur #6 Mrk Mirtnik, Ph.D. Prfssr Th Univrsity f Dlwr mil: mirtni@c.udl.du Wv Prpgtin nd Plritin TM: Trnsvrs lctrmgntic Wvs A md is prticulr fild cnfigurtin. Fr givn lctrmgntic bundry vlu prblm,

More information

Graphs. Graphs. Graphs: Basic Terminology. Directed Graphs. Dr Papalaskari 1

Graphs. Graphs. Graphs: Basic Terminology. Directed Graphs. Dr Papalaskari 1 CSC 00 Disrt Struturs : Introuon to Grph Thory Grphs Grphs CSC 00 Disrt Struturs Villnov Univrsity Grphs r isrt struturs onsisng o vrs n gs tht onnt ths vrs. Grphs n us to mol: omputr systms/ntworks mthml

More information

ME 522 PRINCIPLES OF ROBOTICS. FIRST MIDTERM EXAMINATION April 19, M. Kemal Özgören

ME 522 PRINCIPLES OF ROBOTICS. FIRST MIDTERM EXAMINATION April 19, M. Kemal Özgören ME 522 PINCIPLES OF OBOTICS FIST MIDTEM EXAMINATION April 9, 202 Nm Lst Nm M. Kml Özgörn 2 4 60 40 40 0 80 250 USEFUL FOMULAS cos( ) cos cos sin sin sin( ) sin cos cos sin sin y/ r, cos x/ r, r 0 tn 2(

More information

Factors Success op Ten Critical T the exactly what wonder may you referenced, being questions different the all With success critical ten top the of l

Factors Success op Ten Critical T the exactly what wonder may you referenced, being questions different the all With success critical ten top the of l Fr Su p T rl T xl r rr, bg r ll Wh u rl p l Fllg ll r lkg plr plr rl r kg: 1 k r r u v P 2 u l r P 3 ) r rl k 4 k rprl 5 6 k prbl lvg hkg rl 7 lxbl F 8 l S v 9 p rh L 0 1 k r T h r S pbl r u rl bv p p

More information

(2) If we multiplied a row of B by λ, then the value is also multiplied by λ(here lambda could be 0). namely

(2) If we multiplied a row of B by λ, then the value is also multiplied by λ(here lambda could be 0). namely . DETERMINANT.. Dtrminnt. Introution:I you think row vtor o mtrix s oorint o vtors in sp, thn th gomtri mning o th rnk o th mtrix is th imnsion o th prlllppi spnn y thm. But w r not only r out th imnsion,

More information

² Ý ² ª ² Þ ² Þ Ң Þ ² Þ. ² à INTROIT. huc. per. xi, sti. su- sur. sum, cum. ia : ia, ia : am, num. VR Mi. est. lis. sci. ia, cta. ia.

² Ý ² ª ² Þ ² Þ Ң Þ ² Þ. ² à INTROIT. huc. per. xi, sti. su- sur. sum, cum. ia : ia, ia : am, num. VR Mi. est. lis. sci. ia, cta. ia. str Dy Ps. 138 R 7 r r x, t huc t m m, l : p - í pr m m num m, l l : VR M rá s f ct st sc n -, l l -. Rpt nphn s fr s VR ftr ch vrs Ps. 1. D n, pr bá m, t c g ví m : c g ví ss s nm m m, t r r r c nm m

More information

How much air is required by the people in this lecture theatre during this lecture?

How much air is required by the people in this lecture theatre during this lecture? 3 NTEGRATON tgrtio is us to swr qustios rltig to Ar Volum Totl qutity such s: Wht is th wig r of Boig 747? How much will this yr projct cost? How much wtr os this rsrvoir hol? How much ir is rquir y th

More information

Exhibit 2-9/30/15 Invoice Filing Page 1841 of Page 3660 Docket No

Exhibit 2-9/30/15 Invoice Filing Page 1841 of Page 3660 Docket No xhibit 2-9/3/15 Invie Filing Pge 1841 f Pge 366 Dket. 44498 F u v 7? u ' 1 L ffi s xs L. s 91 S'.e q ; t w W yn S. s t = p '1 F? 5! 4 ` p V -', {} f6 3 j v > ; gl. li -. " F LL tfi = g us J 3 y 4 @" V)

More information

Introduction to Laplace Transforms October 25, 2017

Introduction to Laplace Transforms October 25, 2017 Iroduco o Lplc Trform Ocobr 5, 7 Iroduco o Lplc Trform Lrr ro Mchcl Egrg 5 Smr Egrg l Ocobr 5, 7 Oul Rvw l cl Wh Lplc rform fo of Lplc rform Gg rform b gro Fdg rform d vr rform from bl d horm pplco o dffrl

More information