Fundamentals of Speech Recognition Suggested Project The Hidden Markov Model

Size: px
Start display at page:

Download "Fundamentals of Speech Recognition Suggested Project The Hidden Markov Model"

Transcription

1 . Projec Iroduco Fudameals of Speech Recogo Suggesed Projec The Hdde Markov Model For hs projec, s proposed ha you desg ad mpleme a hdde Markov model (HMM) ha opmally maches he behavor of a se of rag sequeces ha wll be provded o you as par of hs projec. The goal wll be o use he sadard se of forward-backward esmao algorhms o opmally deerme he bes (maxmum lkelhood) HMM ha maches a gve se of rag daa. You are also asked o use he Verb algorhm for esmag he model parameers ad comparg ad corasg he resuls for he wo mehods. You may also wa o vesgae he effecs of usg subses of he rag daa o he esmaed models. Fally, f me perms, you are asked o desg your ow sequece geeraor ad deerme he effecs of chagg he rag sequece characerscs o he esmaed models. 2. HMM Geerave Models A oal of four models were creaed wh he followg characerscs: Model : umber of saes, Q=5 ype of desy of observaos: dscree umber of observao possbles, K=5 ype of model: ergodc sae raso marx desy: radom sae observao marx desy: radom sae pror desy: radom Model 2: umber of saes, Q=5 ype of desy of observaos: dscree umber of observao possbles, K=5 ype of model: ergodc sae raso marx desy: skewed sae observao marx desy: skewed sae pror desy: skewed Model 3: umber of saes, Q=5 ype of desy of observaos: dscree umber of observao possbles, K=5 ype of model: lef-rgh sae raso marx desy: cosraed 2/9/2005 :55 AM HMM Specal Projec

2 sae observao marx desy: radom sae pror desy: cosraed Model 4: umber of saes, Q=5 ype of desy of observaos: dscree umber of observao possbles, K=5 ype of model: lef-rgh sae raso marx desy: cosraed sae observao marx desy: skewed sae pror desy: cosraed A plo showg he geerc srucure of boh he ergodc model (Fgure ) ad he lefrgh model (Fgure 2) s gve below. (Noe ha oly a subse of he sae rasos are show he fgure sce go very messy showg all possble sae-sae raso pahs.) Fgure 5 sae ergodc model (oly some of he acual sae rasos show fgure) Fgure 2 5 sae lef-rgh model 3. HMM Trag Sequeces For hs projec here are four rag ses (avalable from he course webse), labeled:. hmm_observaos_ergodc_radom.ma, 2. hmm_observaos_ergodc_skewed.ma, 2/9/2005 :55 AM 2 HMM Specal Projec

3 3. hmm_observaos_lef-r_radom.ma 4. hmm_observaos_lef-r_skewed.ma The characerscs of hese four daases are as follows:. each ma fle coas he followg formao: a. a observao array, called daa(ex,t), where ex s 50 ad T s 00. The elemes of daa are he observao values (bewee ad K) of each of he observao sequeces as draw from he dscree sae observao marx (see below) b. a hdde saes array, called saes(ex,t). The elemes of saes are he sae dces (bewee ad Q) of each of he observao sequeces as draw from he dscree sae pror desy ad he sae raso marx. The hdde saes array cao be used ay calculao, s provded for say checks o he sysem. c. a durao array, called durao(ex), whch specfes he acual legh of each of he ex rag sequeces. I should be oed ha for he ergodc sequeces, he durao of each sequece s exacly T=00 observaos (wh he acual observaos gve daa), bu for he lef-rgh models, he durao of each sequece s dffere ad s specfed he durao array d. a sae pror array, called pror0(q), whch specfes he probably of each rag sequece sarg each of he Q saes e. a sae raso marx, called rasma0(q,q), whch specfes he probably, a j, of a sae raso from sae o sae j, where <=,j<=q f. a sae observao marx, called obsma0(q,k), whch specfes he probably of symbol k appearg sae, where <=k<=k, ad <=<=Q 2. he frs wo ma fles coa rag sequeces for he ergodc model wh eher radom eres for he sae pror desy, he sae raso marx, ad he sae observao marx (he frs ma fle), or skewed eres for he sae pror desy, he sae raso marx, ad he sae observao marx (he secod ma fle). The secod wo ma fles coa rag sequeces for he lef-rgh model wh eher radom eres for he sae observao marx (he hrd ma fle), or skewed eres for he sae observao marx (he fourh ma fle). The model sae pror desy ad sae raso marx are boh hghly skewed for boh he hrd ad fourh ma fles. Aga s oed ha for lef-rgh models, he durao of he rag sequeces are varable ad are specfed he durao array. For he ergodc models, he durao of each rag sequece s fxed a T=00 observaos. 4. Revew of he Forward-Backward Algorhm The goal of he forward-backward algorhm s o deerme he values of he sae pror desy, he sae raso marx, ad he sae observao desy ha maxmzes he 2/9/2005 :55 AM 3 HMM Specal Projec

4 probably of he rag se. Ths maxmzao s acheved a erave maer, amely by frs posulag a (radom or well educaed) guess as o he values of he sae pror desy, he sae raso marx, ad he sae observao desy, ad he calculag he forward probables, α (, ), for me, sae, ad rag sequece, he backward probables, β (, ), for me, sae, ad rag sequece, he scalg sequece c (, ), for me, ad rag sequece, ad he scaled values of α ad β, amely ˆα ad, ˆ β whch are all compued as follows:. Oba al esmaes of he sae pror desy, he sae raso marx, ad he sae observao marx from eher radom guesses, or from skewed al codos. Ial esmaes of π, a, ad b of he form: π = { π, π,..., π }; π = probably of beg sae a me 2 Q a a2 a Q a2 a22 a 2Q a= aq aq2 a QQ aj = probably of makg a raso from sae o sae j b() b(2) b( K) b2() b2(2) b2( k) b = bq() bq(2) bq( K) h b( k) = probably of observg he k symbol sae 2. Use as he rag se oe of he se of observaos read earler. We deoe he complee rag se of ex sequeces of observaos, wh T beg he durao of he -h observao sequece, as: O = value of he observao a me ad for sequece, O K, =,2,..., T, =,2,..., ex 2/9/2005 :55 AM 4 HMM Specal Projec

5 3. Compue he forward, backward, scale facor, ad scaled esmaes of α ad β as: Forward Loop--For each observao sequece =, 2,..., ex he rag se (wh sequece durao T ), compue he followg: Ialzao Sep ( = ) : α (, ) = π b( O), =, 2,..., Q(uscaled α) Q c(, ) = α (, ) = = ˆ α (, ) = α (, )/ c(, ) (scaled α) Ierao Sep: for = 2,3,..., T Q α (, ) ˆ = α ( ja, ) j b( O ), =, 2,..., Q j= Q c (, ) = α (, ) (scale facors) ˆ α (, ) = α (, ) / c (, ), =, 2,..., Q Backward Loop--For each observao sequece =, 2,..., ex he rag se (wh sequece durao T ), compue he followg: Ialzao Sep: β (, ) = ; =, 2,..., Q T ˆ β (, ) = ; =,2,..., Q T Ierao Sep: for = T, T 2,..., Q ˆ j β+ j + j= (uscaled β) (scaled β ) β (, ) = a ( jb, ) ( O ), =, 2,..., Q ˆ β (, ) = β (, ) / c ( +, ), =, 2,..., Q 2/9/2005 :55 AM 5 HMM Specal Projec

6 4. Compue he sae desy, γ (, ), ad he sae raso desy, ξ (, j, ), as: γ (, ) = probably of beg sae a me for sequece γ (, ) ˆ (, ) ˆ = α β(, ), =, 2,..., Q, =, 2,..., T, =, 2,..., ex ξ (, j, ) = probably of beg sae a me ad makg a raso o sae j a me + for sequece ξ (,, ) ˆ (, ) ( ˆ j = α aj bj O+ ) β+ ( j, )/ c( +, ), =,2,..., Q, j =, 2,..., Q, =, 2,..., T, =,2,..., ex 5. Re-esmae he sae pror desy, he sae raso marx ad he sae observao marx, usg he relaos:. re-esmao of sae pror desy: ex γ(, ) = π =, =,2,..., Q ex 2. re-esmao of sae raso marx: ex T ξ (, j, ) = = aj =, =, 2,..., Q, j =, 2,..., Q ex T γ (, ) = = 3. re-esmao of sae observao desy marx: ex T = = O = k = = γ ( j, ) bj ( k) =, j =, 2,..., Q, k =,2,..., K ex T γ ( j, ) 6. Trag sequece lkelhood calculao. I s esseal whe dog re-esmao of HMM parameers o moor he egave log probably of he oal se of observao sequeces o esure ha a each erao he oal egave log probably s decreasg, 2/9/2005 :55 AM 6 HMM Specal Projec

7 ad as a check as o how closely he lkelhoods from he bes esmae mach or exceed he lkelhood from he acual geeraor deses. Ths calculao of oal egave log probably ca be accomplshed by keepg rack, a each erao, of he sum of he egave logs of all he scale facors,.e., L rag ex T = = = log( c (, )) A eresg feaure ha ca be used whe he acual geeraor sequece s kow s o do a sgle forward-backward pass usg he acual geeraor sae pror desy, he acual sae raso marx, ad he acual sae observao marx, hereby gvg he rag sequece lkelhood for he acual geeraor. Ieresgly, he bes esmaed model ca have a rag sequece lkelhood ha s acually hgher ha ha of he geeraor sequece especally whe he acual sequeces are geeraed from radomly geeraed sae prors, sae raso marces, ad sae observao marces. However, mos of he me, he rag sequece lkelhoods are worse ha ha of he geeraor sequece a leas ul he re-esmao procedure has reached a sable opmum ad o goe suck a a local mmum. 5. Revew of he Verb Algorhm A alerave approach o he forward-backward mehod, for re-esmao of HMM model parameers, s he Verb algorhm whch herely fds he bes machg pah ha bes algs each rag sequece wh he curre model. By keepg rack oly of he bes sequece, he compuao ca be reduced cosderably. However, cases where he bes pah s o much beer ha alerave pahs, he Verb algorhm ca lead o hghly sub-opmal soluos. Ths occurs more ofe wh radomly creaed HMM model parameers ha wh skewed model parameers, or wh lef-rgh models whch have srog cosras o he sae pror desy ad he sae raso desy. The deals of how he Verb algorhm operaes are as follows: 2/9/2005 :55 AM 7 HMM Specal Projec

8 . Coverso o log probables--oe me operao π = log( π ), =, 2,..., Q a = log( a ), =,2,..., Q, j =,2,..., Q j j b ( k) = log( b( k)), =,2,..., Q, k =,2,..., K 2. For each sequece =,2,..., ex he rag se (wh sequece durao T ), compue he followg: Ialzao ( = ): δ(, ) = π + b ( O ), ψ(, ) = 0; =, 2,..., Q =,2,..., Q Recurso: for = 2,3,..., T δ (, ) max (, ) ( O ), j =, 2,..., Q j = δ + aj + bj Q ψ ( j, ) = arg max δ (, ) a +, j=,2,..., Q Termao: j Q P ( ) = max δt (, ) Q qt ( ) = argmax δt (, ) Q Backrackg he bes sae sequece: for = T, T 2,..., q =ψ + + ( ) ( q ( )) Oce he Verb algme of he rag sequeces wh he curre model has bee obaed, we have esseally ucovered he hdde pars of he model, so we have a uque assgme of each observao o a dsc model sae. Hece he re-esmao of he sae pror desy, he sae raso marx, ad he sae observao marx s cosderably smpler ha for he forward-backward mehod, ad ca be saed smply as follows: 2/9/2005 :55 AM 8 HMM Specal Projec

9 . re-esmao of sae pror desy: ex cq [ ( ) = ] = π = ex, =, 2,..., Q where cq [ ( ) = ] = whe q ( ) = ad s zero oherwse 2. re-esmao of sae raso marx: a j = ex T = = cq [ ( ) = q, + ( ) = j], =, 2,..., Q, j =, 2,..., Q ex T cq [ ( ) = ] = = where c[ q ( ) =, q ( ) = j] = whe q ( ) = ad q ( ) = j + + ad s zero oherwse 3. re-esmao of sae observao desy marx: ex T cq [ ( ) = jo, = k] = = j ( ) =, =,2,...,, =,2,..., ex T cq [ ( ) = j] = = b k j Q k K where [ ( ), ] whe ( ) ad cq = jo = k = q = j O = k ad s zero oherwse 6. HMM Esmao Projec The purpose of hs projec s o each you how o realze boh a forward-backward ad a Verb procedure for esmag parameers of Hdde Markov Models. To ha ed you are asked o read each of he four rag ses, ad o program HMM re-esmao algorhms usg boh a forward-backward approach ad a Verb approach, ad o compare ad coras he wo procedures erms of speed, accuracy, sesvy o al esmaes, compuaoal ssues (akg logs of zero-valued quaes), ad ay oher ssue ha arses durg he course of hs projec. 2/9/2005 :55 AM 9 HMM Specal Projec

10 I s suggesed ha you frs wre some smple code o geerae a radom sae pror desy, a radom sae raso desy, ad a radom sae observao desy for he parameers of hs exercse, amely Q=5 sae models, wh K=5 possble observaos each sae. The acual geeraor deses are labeled pror0, rasma0 ad obsma0 so you mgh wa o label your radom selecos as pror, rasma, ad obsma o dsgush hem from he acual sequece geeraors. Oce you read each of he rag ses, famlarze yourself wh he daa sequece whch s of he form daa(ex,t) where ex s 50 ad T s 00. Ths daa sequece s he observed se of oupus of he model, bu, of course, here are o observed saes as he saes are hdde. Acually he array saes(ex,t) has he geeraor sae daa bu hs cao be used ay maer your smulaos. You should also remd yourself ha he durao of each rag sequece (=,2,,ex) s specfed he array durao(:ex). (For he ergodc models ha we wll be usg, he durao of each sequece s T=00. For he lef-rgh models he durao s varable ad s specfed durao(:ex).) Oce he rag daa has bee read, he frs sep you should do s score he rag se o he geeraor model by rug oe full erao of he forward-backward roue (whch you have o wre), ad by og he sum of he egave logs of he scalg sequeces. Ths lkelhood score represes a arge value for your acual re-esmao roue a arge ha you wll somemes surpass slghly (why s hs he case) ad mosly wll mss hg due o local opma or bad al sars. The ex sep s o creae full forward-backward esmao ad HMM model reesmao roues, as well as a Verb esmao ad HMM model re-esmao roue. Oce hese roues have bee debugged, you ca beg playg wh he four rag ses ad deermg how well you ca esmae he model parameers. I s a eresg exercse o use he acual HMM deses (raher ha radom esmaes) o deerme how he oal log lkelhood score vares as you re-esmae he deses sarg from he deal codos ; however hs exercse s jus o show how good a soluo ca be obaed f you ca ge pas local mma of he re-esmao roues. I summary, you are asked o buld wo ses of HMM re-esmao roues, oe based o he forward-backward algorhm, oe based o he Verb sae sequece, ad o deerme how well your algorhms work o wo ypes of models, amely a ergodc model wh 5 saes ad 5 possble observaos each sae, ad a lef-rgh model wh 5 saes ad 5 possble observaos each sae. You should cosder usg several radom sars o deerme he bes soluo for each model ad each rag se. You are also gve wo versos of each model, oe wh radomly chose values for he sae pror desy, he sae raso marx ad he sae observao marx, ad oe wh skewed values for he sae pror desy ad he sae raso marx. Hece you should experme wh each of hese rag ses o udersad whch rag mehods work bes, ad why. Fally, you should vesgae he effecs of usg hghly cosraed models (such as he lef-rgh model) o he sae pror desy esmao ad he sae raso marx esmao. 2/9/2005 :55 AM 0 HMM Specal Projec

11 Oher quesos you should address as par of hs projec clude he followg:. for cases whe he coverged lkelhood s comparable o he geeraor lkelhood, how do he model deses mach hose of he geeraor? 2. for cases whe he coverged lkelhood s comparable o he geeraor lkelhood, how do he model saes mach hose of he geeraor? 3. for cases whe he coverged lkelhood s smaller ha he geeraor lkelhood (.e., covergece a a local mmum bu o he global mmum), wha s happeg wh he saes ad how do he resulg models compare o he geeraor model? 4. how does he speed of covergece compare for he forward-backward re-esmao ad he Verb re-esmao mehods. 5. how do he compuaoal requremes for he forward-backward ad Verb mehods compare? 6. how do he lkelhood scores of he geeraor models compare whe usg forwardbackward scorg wh hose whe usg Verb scorg? Wha accous for hese dffereces ad how do hey compare for radom ad skewed rag sequeces. 7. wha do you hk would be he effec of fewer rag sequeces (you ca ry hs ou by usg less ha he ex=50 sequeces suppled)? Wha do you hk would be he effec of more rag sequeces? If you are successful makg he re-esmao mehods work properly ad effcely, you mgh wa o buld your ow HMM sequece geeraor ad experme wh creag a rage of model observaos ad see how well your re-esmao roues work o hs ew daa. I s especally eresg o vesgae he effec of creased umbers of observao sequeces o deerme how may are eeded o make he covergece more rapd ad less sesve o he al esmaes of model parameers. 2/9/2005 :55 AM HMM Specal Projec

14. Poisson Processes

14. Poisson Processes 4. Posso Processes I Lecure 4 we roduced Posso arrvals as he lmg behavor of Bomal radom varables. Refer o Posso approxmao of Bomal radom varables. From he dscusso here see 4-6-4-8 Lecure 4 " arrvals occur

More information

Continuous Time Markov Chains

Continuous Time Markov Chains Couous me Markov chas have seay sae probably soluos f a oly f hey are ergoc, us lke scree me Markov chas. Fg he seay sae probably vecor for a couous me Markov cha s o more ffcul ha s he scree me case,

More information

Least squares and motion. Nuno Vasconcelos ECE Department, UCSD

Least squares and motion. Nuno Vasconcelos ECE Department, UCSD Leas squares ad moo uo Vascocelos ECE Deparme UCSD Pla for oda oda we wll dscuss moo esmao hs s eresg wo was moo s ver useful as a cue for recogo segmeao compresso ec. s a grea eample of leas squares problem

More information

FALL HOMEWORK NO. 6 - SOLUTION Problem 1.: Use the Storage-Indication Method to route the Input hydrograph tabulated below.

FALL HOMEWORK NO. 6 - SOLUTION Problem 1.: Use the Storage-Indication Method to route the Input hydrograph tabulated below. Jorge A. Ramírez HOMEWORK NO. 6 - SOLUTION Problem 1.: Use he Sorage-Idcao Mehod o roue he Ipu hydrograph abulaed below. Tme (h) Ipu Hydrograph (m 3 /s) Tme (h) Ipu Hydrograph (m 3 /s) 0 0 90 450 6 50

More information

Least Squares Fitting (LSQF) with a complicated function Theexampleswehavelookedatsofarhavebeenlinearintheparameters

Least Squares Fitting (LSQF) with a complicated function Theexampleswehavelookedatsofarhavebeenlinearintheparameters Leas Squares Fg LSQF wh a complcaed fuco Theeampleswehavelookedasofarhavebeelearheparameers ha we have bee rg o deerme e.g. slope, ercep. For he case where he fuco s lear he parameers we ca fd a aalc soluo

More information

Solution set Stat 471/Spring 06. Homework 2

Solution set Stat 471/Spring 06. Homework 2 oluo se a 47/prg 06 Homework a Whe he upper ragular elemes are suppressed due o smmer b Le Y Y Y Y A weep o he frs colum o oba: A ˆ b chagg he oao eg ad ec YY weep o he secod colum o oba: Aˆ YY weep o

More information

Determination of Antoine Equation Parameters. December 4, 2012 PreFEED Corporation Yoshio Kumagae. Introduction

Determination of Antoine Equation Parameters. December 4, 2012 PreFEED Corporation Yoshio Kumagae. Introduction refeed Soluos for R&D o Desg Deermao of oe Equao arameers Soluos for R&D o Desg December 4, 0 refeed orporao Yosho Kumagae refeed Iroduco hyscal propery daa s exremely mpora for performg process desg ad

More information

QR factorization. Let P 1, P 2, P n-1, be matrices such that Pn 1Pn 2... PPA

QR factorization. Let P 1, P 2, P n-1, be matrices such that Pn 1Pn 2... PPA QR facorzao Ay x real marx ca be wre as AQR, where Q s orhogoal ad R s upper ragular. To oba Q ad R, we use he Householder rasformao as follows: Le P, P, P -, be marces such ha P P... PPA ( R s upper ragular.

More information

(1) Cov(, ) E[( E( ))( E( ))]

(1) Cov(, ) E[( E( ))( E( ))] Impac of Auocorrelao o OLS Esmaes ECON 3033/Evas Cosder a smple bvarae me-seres model of he form: y 0 x The four key assumpos abou ε hs model are ) E(ε ) = E[ε x ]=0 ) Var(ε ) =Var(ε x ) = ) Cov(ε, ε )

More information

Real-time Classification of Large Data Sets using Binary Knapsack

Real-time Classification of Large Data Sets using Binary Knapsack Real-me Classfcao of Large Daa Ses usg Bary Kapsack Reao Bru bru@ds.uroma. Uversy of Roma La Sapeza AIRO 004-35h ANNUAL CONFERENCE OF THE ITALIAN OPERATIONS RESEARCH Sepember 7-0, 004, Lecce, Ialy Oule

More information

Cyclone. Anti-cyclone

Cyclone. Anti-cyclone Adveco Cycloe A-cycloe Lorez (963) Low dmesoal aracors. Uclear f hey are a good aalogy o he rue clmae sysem, bu hey have some appealg characerscs. Dscusso Is he al codo balaced? Is here a al adjusme

More information

The Poisson Process Properties of the Poisson Process

The Poisson Process Properties of the Poisson Process Posso Processes Summary The Posso Process Properes of he Posso Process Ierarrval mes Memoryless propery ad he resdual lfeme paradox Superposo of Posso processes Radom seleco of Posso Pos Bulk Arrvals ad

More information

The ray paths and travel times for multiple layers can be computed using ray-tracing, as demonstrated in Lab 3.

The ray paths and travel times for multiple layers can be computed using ray-tracing, as demonstrated in Lab 3. C. Trael me cures for mulple reflecors The ray pahs ad rael mes for mulple layers ca be compued usg ray-racg, as demosraed Lab. MATLAB scrp reflec_layers_.m performs smple ray racg. (m) ref(ms) ref(ms)

More information

-distributed random variables consisting of n samples each. Determine the asymptotic confidence intervals for

-distributed random variables consisting of n samples each. Determine the asymptotic confidence intervals for Assgme Sepha Brumme Ocober 8h, 003 9 h semeser, 70544 PREFACE I 004, I ed o sped wo semesers o a sudy abroad as a posgraduae exchage sude a he Uversy of Techology Sydey, Ausrala. Each opporuy o ehace my

More information

Midterm Exam. Tuesday, September hour, 15 minutes

Midterm Exam. Tuesday, September hour, 15 minutes Ecoomcs of Growh, ECON560 Sa Fracsco Sae Uvers Mchael Bar Fall 203 Mderm Exam Tuesda, Sepember 24 hour, 5 mues Name: Isrucos. Ths s closed boo, closed oes exam. 2. No calculaors of a d are allowed. 3.

More information

8. Queueing systems lect08.ppt S Introduction to Teletraffic Theory - Fall

8. Queueing systems lect08.ppt S Introduction to Teletraffic Theory - Fall 8. Queueg sysems lec8. S-38.45 - Iroduco o Teleraffc Theory - Fall 8. Queueg sysems Coes Refresher: Smle eleraffc model M/M/ server wag laces M/M/ servers wag laces 8. Queueg sysems Smle eleraffc model

More information

Real-Time Systems. Example: scheduling using EDF. Feasibility analysis for EDF. Example: scheduling using EDF

Real-Time Systems. Example: scheduling using EDF. Feasibility analysis for EDF. Example: scheduling using EDF EDA/DIT6 Real-Tme Sysems, Chalmers/GU, 0/0 ecure # Updaed February, 0 Real-Tme Sysems Specfcao Problem: Assume a sysem wh asks accordg o he fgure below The mg properes of he asks are gve he able Ivesgae

More information

Learning of Graphical Models Parameter Estimation and Structure Learning

Learning of Graphical Models Parameter Estimation and Structure Learning Learg of Grahal Models Parameer Esmao ad Sruure Learg e Fukumzu he Isue of Sasal Mahemas Comuaoal Mehodology Sasal Iferee II Work wh Grahal Models Deermg sruure Sruure gve by modelg d e.g. Mxure model

More information

AML710 CAD LECTURE 12 CUBIC SPLINE CURVES. Cubic Splines Matrix formulation Normalised cubic splines Alternate end conditions Parabolic blending

AML710 CAD LECTURE 12 CUBIC SPLINE CURVES. Cubic Splines Matrix formulation Normalised cubic splines Alternate end conditions Parabolic blending CUIC SLINE CURVES Cubc Sples Marx formulao Normalsed cubc sples Alerae ed codos arabolc bledg AML7 CAD LECTURE CUIC SLINE The ame sple comes from he physcal srume sple drafsme use o produce curves A geeral

More information

Speech, NLP and the Web

Speech, NLP and the Web peech NL ad he Web uhpak Bhaacharyya CE Dep. IIT Bombay Lecure 38: Uuperved learg HMM CFG; Baum Welch lecure 37 wa o cogve NL by Abh Mhra Baum Welch uhpak Bhaacharyya roblem HMM arg emac ar of peech Taggg

More information

Lecture 3 Topic 2: Distributions, hypothesis testing, and sample size determination

Lecture 3 Topic 2: Distributions, hypothesis testing, and sample size determination Lecure 3 Topc : Drbuo, hypohe eg, ad ample ze deermao The Sude - drbuo Coder a repeaed drawg of ample of ze from a ormal drbuo of mea. For each ample, compue,,, ad aoher ac,, where: The ac he devao of

More information

As evident from the full-sample-model, we continue to assume that individual errors are identically and

As evident from the full-sample-model, we continue to assume that individual errors are identically and Maxmum Lkelhood smao Greee Ch.4; App. R scrp modsa, modsb If we feel safe makg assumpos o he sascal dsrbuo of he error erm, Maxmum Lkelhood smao (ML) s a aracve alerave o Leas Squares for lear regresso

More information

θ = θ Π Π Parametric counting process models θ θ θ Log-likelihood: Consider counting processes: Score functions:

θ = θ Π Π Parametric counting process models θ θ θ Log-likelihood: Consider counting processes: Score functions: Paramerc coug process models Cosder coug processes: N,,..., ha cou he occurreces of a eve of eres for dvduals Iesy processes: Lelhood λ ( ;,,..., N { } λ < Log-lelhood: l( log L( Score fucos: U ( l( log

More information

Linear Regression Linear Regression with Shrinkage

Linear Regression Linear Regression with Shrinkage Lear Regresso Lear Regresso h Shrkage Iroduco Regresso meas predcg a couous (usuall scalar oupu from a vecor of couous pus (feaures x. Example: Predcg vehcle fuel effcec (mpg from 8 arbues: Lear Regresso

More information

CS344: Introduction to Artificial Intelligence

CS344: Introduction to Artificial Intelligence C344: Iroduco o Arfcal Iellgece Puhpa Bhaacharyya CE Dep. IIT Bombay Lecure 3 3 32 33: Forward ad bacward; Baum elch 9 h ad 2 March ad 2 d Aprl 203 Lecure 27 28 29 were o EM; dae 2 h March o 8 h March

More information

The Linear Regression Of Weighted Segments

The Linear Regression Of Weighted Segments The Lear Regresso Of Weghed Segmes George Dael Maeescu Absrac. We proposed a regresso model where he depede varable s made o up of pos bu segmes. Ths suao correspods o he markes hroughou he da are observed

More information

Final Exam Applied Econometrics

Final Exam Applied Econometrics Fal Eam Appled Ecoomercs. 0 Sppose we have he followg regresso resl: Depede Varable: SAT Sample: 437 Iclded observaos: 437 Whe heeroskedasc-cosse sadard errors & covarace Varable Coeffce Sd. Error -Sasc

More information

Partial Molar Properties of solutions

Partial Molar Properties of solutions Paral Molar Properes of soluos A soluo s a homogeeous mxure; ha s, a soluo s a oephase sysem wh more ha oe compoe. A homogeeous mxures of wo or more compoes he gas, lqud or sold phase The properes of a

More information

Continuous Indexed Variable Systems

Continuous Indexed Variable Systems Ieraoal Joural o Compuaoal cece ad Mahemacs. IN 0974-389 Volume 3, Number 4 (20), pp. 40-409 Ieraoal Research Publcao House hp://www.rphouse.com Couous Idexed Varable ysems. Pouhassa ad F. Mohammad ghjeh

More information

Solution. The straightforward approach is surprisingly difficult because one has to be careful about the limits.

Solution. The straightforward approach is surprisingly difficult because one has to be careful about the limits. ose ad Varably Homewor # (8), aswers Q: Power spera of some smple oses A Posso ose A Posso ose () s a sequee of dela-fuo pulses, eah ourrg depedely, a some rae r (More formally, s a sum of pulses of wdh

More information

Quantum Mechanics II Lecture 11 Time-dependent perturbation theory. Time-dependent perturbation theory (degenerate or non-degenerate starting state)

Quantum Mechanics II Lecture 11 Time-dependent perturbation theory. Time-dependent perturbation theory (degenerate or non-degenerate starting state) Pro. O. B. Wrgh, Auum Quaum Mechacs II Lecure Tme-depede perurbao heory Tme-depede perurbao heory (degeerae or o-degeerae sarg sae) Cosder a sgle parcle whch, s uperurbed codo wh Hamloa H, ca exs a superposo

More information

Optimal Eye Movement Strategies in Visual Search (Supplement)

Optimal Eye Movement Strategies in Visual Search (Supplement) Opmal Eye Moveme Sraeges Vsual Search (Suppleme) Jr Naemk ad Wlso S. Gesler Ceer for Percepual Sysems ad Deparme of Psychology, Uversy of exas a Aus, Aus X 787 Here we derve he deal searcher for he case

More information

Suppose we have observed values t 1, t 2, t n of a random variable T.

Suppose we have observed values t 1, t 2, t n of a random variable T. Sppose we have obseved vales, 2, of a adom vaable T. The dsbo of T s ow o belog o a cea ype (e.g., expoeal, omal, ec.) b he veco θ ( θ, θ2, θp ) of ow paamees assocaed wh s ow (whee p s he mbe of ow paamees).

More information

International Journal Of Engineering And Computer Science ISSN: Volume 5 Issue 12 Dec. 2016, Page No.

International Journal Of Engineering And Computer Science ISSN: Volume 5 Issue 12 Dec. 2016, Page No. www.jecs. Ieraoal Joural Of Egeerg Ad Compuer Scece ISSN: 19-74 Volume 5 Issue 1 Dec. 16, Page No. 196-1974 Sofware Relably Model whe mulple errors occur a a me cludg a faul correco process K. Harshchadra

More information

Chapter 3: Maximum-Likelihood & Bayesian Parameter Estimation (part 1)

Chapter 3: Maximum-Likelihood & Bayesian Parameter Estimation (part 1) Aoucemes Reags o E-reserves Proec roosal ue oay Parameer Esmao Bomercs CSE 9-a Lecure 6 CSE9a Fall 6 CSE9a Fall 6 Paer Classfcao Chaer 3: Mamum-Lelhoo & Bayesa Parameer Esmao ar All maerals hese sles were

More information

Key words: Fractional difference equation, oscillatory solutions,

Key words: Fractional difference equation, oscillatory solutions, OSCILLATION PROPERTIES OF SOLUTIONS OF FRACTIONAL DIFFERENCE EQUATIONS Musafa BAYRAM * ad Ayd SECER * Deparme of Compuer Egeerg, Isabul Gelsm Uversy Deparme of Mahemacal Egeerg, Yldz Techcal Uversy * Correspodg

More information

EE 6885 Statistical Pattern Recognition

EE 6885 Statistical Pattern Recognition EE 6885 Sascal Paer Recogo Fall 005 Prof. Shh-Fu Chag hp://www.ee.columba.edu/~sfchag Lecure 5 (9//05 4- Readg Model Parameer Esmao ML Esmao, Chap. 3. Mure of Gaussa ad EM Referece Boo, HTF Chap. 8.5 Teboo,

More information

Density estimation III.

Density estimation III. Lecure 4 esy esmao III. Mlos Hauskrec mlos@cs..edu 539 Seo Square Oule Oule: esy esmao: Mamum lkelood ML Bayesa arameer esmaes MP Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Eoeal

More information

Efficient Estimators for Population Variance using Auxiliary Information

Efficient Estimators for Population Variance using Auxiliary Information Global Joural of Mahemacal cece: Theor ad Praccal. IN 97-3 Volume 3, Number (), pp. 39-37 Ieraoal Reearch Publcao Houe hp://www.rphoue.com Effce Emaor for Populao Varace ug Aular Iformao ubhah Kumar Yadav

More information

Asymptotic Behavior of Solutions of Nonlinear Delay Differential Equations With Impulse

Asymptotic Behavior of Solutions of Nonlinear Delay Differential Equations With Impulse P a g e Vol Issue7Ver,oveber Global Joural of Scece Froer Research Asypoc Behavor of Soluos of olear Delay Dffereal Equaos Wh Ipulse Zhag xog GJSFR Classfcao - F FOR 3 Absrac Ths paper sudes he asypoc

More information

VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS. Hunan , China,

VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS. Hunan , China, Mahemacal ad Compuaoal Applcaos Vol. 5 No. 5 pp. 834-839. Assocao for Scefc Research VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS Hoglag Lu Aguo Xao Yogxag Zhao School of Mahemacs

More information

General Complex Fuzzy Transformation Semigroups in Automata

General Complex Fuzzy Transformation Semigroups in Automata Joural of Advaces Compuer Research Quarerly pissn: 345-606x eissn: 345-6078 Sar Brach Islamc Azad Uversy Sar IRIra Vol 7 No May 06 Pages: 7-37 wwwacrausaracr Geeral Complex uzzy Trasformao Semgroups Auomaa

More information

Available online Journal of Scientific and Engineering Research, 2014, 1(1): Research Article

Available online  Journal of Scientific and Engineering Research, 2014, 1(1): Research Article Avalable ole wwwjsaercom Joural o Scec ad Egeerg Research, 0, ():0-9 Research Arcle ISSN: 39-630 CODEN(USA): JSERBR NEW INFORMATION INEUALITIES ON DIFFERENCE OF GENERALIZED DIVERGENCES AND ITS APPLICATION

More information

FORCED VIBRATION of MDOF SYSTEMS

FORCED VIBRATION of MDOF SYSTEMS FORCED VIBRAION of DOF SSES he respose of a N DOF sysem s govered by he marx equao of moo: ] u C] u K] u 1 h al codos u u0 ad u u 0. hs marx equao of moo represes a sysem of N smulaeous equaos u ad s me

More information

Fault Tolerant Computing. Fault Tolerant Computing CS 530 Probabilistic methods: overview

Fault Tolerant Computing. Fault Tolerant Computing CS 530 Probabilistic methods: overview Probably 1/19/ CS 53 Probablsc mehods: overvew Yashwa K. Malaya Colorado Sae Uversy 1 Probablsc Mehods: Overvew Cocree umbers presece of uceray Probably Dsjo eves Sascal depedece Radom varables ad dsrbuos

More information

Reliability Analysis of Sparsely Connected Consecutive-k Systems: GERT Approach

Reliability Analysis of Sparsely Connected Consecutive-k Systems: GERT Approach Relably Aalyss of Sparsely Coece Cosecuve- Sysems: GERT Approach Pooa Moha RMSI Pv. L Noa-2131 poalovely@yahoo.com Mau Agarwal Deparme of Operaoal Research Uversy of Delh Delh-117, Ia Agarwal_maulaa@yahoo.com

More information

Machine Learning. Hidden Markov Model. Eric Xing / /15-781, 781, Fall Lecture 17, March 24, 2008

Machine Learning. Hidden Markov Model. Eric Xing / /15-781, 781, Fall Lecture 17, March 24, 2008 Mache Learg 0-70/5 70/5-78 78 Fall 2008 Hdde Marov Model Erc Xg Lecure 7 March 24 2008 Readg: Cha. 3 C.B boo Erc Xg Erc Xg 2 Hdde Marov Model: from sac o damc mure models Sac mure Damc mure Y Y Y 2 Y 3

More information

Comparison of the Bayesian and Maximum Likelihood Estimation for Weibull Distribution

Comparison of the Bayesian and Maximum Likelihood Estimation for Weibull Distribution Joural of Mahemacs ad Sascs 6 (2): 1-14, 21 ISSN 1549-3644 21 Scece Publcaos Comarso of he Bayesa ad Maxmum Lkelhood Esmao for Webull Dsrbuo Al Omar Mohammed Ahmed, Hadeel Salm Al-Kuub ad Noor Akma Ibrahm

More information

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square Lecure 5 esy esmao Mlos Hauskrec mlos@cs..edu 539 Seo Square esy esmaos ocs: esy esmao: Mamum lkelood ML Bayesa arameer esmaes M Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Noaramerc

More information

Solution of Impulsive Differential Equations with Boundary Conditions in Terms of Integral Equations

Solution of Impulsive Differential Equations with Boundary Conditions in Terms of Integral Equations Joural of aheacs ad copuer Scece (4 39-38 Soluo of Ipulsve Dffereal Equaos wh Boudary Codos Ters of Iegral Equaos Arcle hsory: Receved Ocober 3 Acceped February 4 Avalable ole July 4 ohse Rabba Depare

More information

Multiple Choice Test. Chapter Adequacy of Models for Regression

Multiple Choice Test. Chapter Adequacy of Models for Regression Multple Choce Test Chapter 06.0 Adequac of Models for Regresso. For a lear regresso model to be cosdered adequate, the percetage of scaled resduals that eed to be the rage [-,] s greater tha or equal to

More information

COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION

COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION Eldesoky E. Affy. Faculy of Eg. Shbee El kom Meoufa Uv. Key word : Raylegh dsrbuo, leas squares mehod, relave leas squares, leas absolue

More information

Cyclically Interval Total Colorings of Cycles and Middle Graphs of Cycles

Cyclically Interval Total Colorings of Cycles and Middle Graphs of Cycles Ope Joural of Dsree Mahemas 2017 7 200-217 hp://wwwsrporg/joural/ojdm ISSN Ole: 2161-7643 ISSN Pr: 2161-7635 Cylally Ierval Toal Colorgs of Cyles Mddle Graphs of Cyles Yogqag Zhao 1 Shju Su 2 1 Shool of

More information

Some Probability Inequalities for Quadratic Forms of Negatively Dependent Subgaussian Random Variables

Some Probability Inequalities for Quadratic Forms of Negatively Dependent Subgaussian Random Variables Joural of Sceces Islamc epublc of Ira 6(: 63-67 (005 Uvers of ehra ISSN 06-04 hp://scecesuacr Some Probabl Iequales for Quadrac Forms of Negavel Depede Subgaussa adom Varables M Am A ozorga ad H Zare 3

More information

The Mean Residual Lifetime of (n k + 1)-out-of-n Systems in Discrete Setting

The Mean Residual Lifetime of (n k + 1)-out-of-n Systems in Discrete Setting Appled Mahemacs 4 5 466-477 Publshed Ole February 4 (hp//wwwscrporg/oural/am hp//dxdoorg/436/am45346 The Mea Resdual Lfeme of ( + -ou-of- Sysems Dscree Seg Maryam Torab Sahboom Deparme of Sascs Scece ad

More information

Brownian Motion and Stochastic Calculus. Brownian Motion and Stochastic Calculus

Brownian Motion and Stochastic Calculus. Brownian Motion and Stochastic Calculus Browa Moo Sochasc Calculus Xogzh Che Uversy of Hawa a Maoa earme of Mahemacs Seember, 8 Absrac Ths oe s abou oob decomoso he bascs of Suare egrable margales Coes oob-meyer ecomoso Suare Iegrable Margales

More information

Economics 8723 Macroeconomic Theory Problem Set 2 Professor Sanjay Chugh Spring 2017

Economics 8723 Macroeconomic Theory Problem Set 2 Professor Sanjay Chugh Spring 2017 Deparme of Ecoomics The Ohio Sae Uiversiy Ecoomics 8723 Macroecoomic Theory Problem Se 2 Professor Sajay Chugh Sprig 207 Labor Icome Taxes, Nash-Bargaied Wages, ad Proporioally-Bargaied Wages. I a ecoomy

More information

IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS

IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS Vol.7 No.4 (200) p73-78 Joural of Maageme Scece & Sascal Decso IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS TIANXIANG YAO AND ZAIWU GONG College of Ecoomcs &

More information

Solving Non-Linear Rational Expectations Models: Approximations based on Taylor Expansions

Solving Non-Linear Rational Expectations Models: Approximations based on Taylor Expansions Work progress Solvg No-Lear Raoal Expecaos Models: Approxmaos based o Taylor Expasos Rober Kollma (*) Deparme of Ecoomcs, Uversy of Pars XII 6, Av. du Gééral de Gaulle; F-94 Créel Cedex; Frace rober_kollma@yahoo.com;

More information

4 5 = So 2. No, as = ± and invariant factor 6. Solution 3 Each of (1, 0),(1, 2),(0, 2) has order 2 and generates a C

4 5 = So 2. No, as = ± and invariant factor 6. Solution 3 Each of (1, 0),(1, 2),(0, 2) has order 2 and generates a C Soluos (page 7) ρ 5 4 4 Soluo = ρ as 4 4 5 = So ρ, ρ s a Z bass of Z ad 5 4 ( m, m ) = (,7) = (,9) No, as 5 7 4 5 6 = ± Soluo The 6 elemes of Z K are: g = K + e + e, g = K + e, g = K + e, 4g = K + e, 5g

More information

Discrete Markov Process. Introduction. Example: Balls and Urns. Stochastic Automaton. INTRODUCTION TO Machine Learning 3rd Edition

Discrete Markov Process. Introduction. Example: Balls and Urns. Stochastic Automaton. INTRODUCTION TO Machine Learning 3rd Edition EHEM ALPAYDI he MI Press, 04 Lecure Sldes for IRODUCIO O Machne Learnng 3rd Edon alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/ml3e Sldes from exboo resource page. Slghly eded and wh addonal examples

More information

Pricing Asian Options with Fourier Convolution

Pricing Asian Options with Fourier Convolution Prcg Asa Opos wh Fourer Covoluo Cheg-Hsug Shu Deparme of Compuer Scece ad Iformao Egeerg Naoal Tawa Uversy Coes. Iroduco. Backgroud 3. The Fourer Covoluo Mehod 3. Seward ad Hodges facorzao 3. Re-ceerg

More information

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model Probablsc Model for Tme-seres Daa: Hdden Markov Model Hrosh Mamsuka Bonformacs Cener Kyoo Unversy Oulne Three Problems for probablsc models n machne learnng. Compung lkelhood 2. Learnng 3. Parsng (predcon

More information

Probability Bracket Notation and Probability Modeling. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA 94087, USA. Abstract

Probability Bracket Notation and Probability Modeling. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA 94087, USA. Abstract Probably Bracke Noao ad Probably Modelg Xg M. Wag Sherma Vsual Lab, Suyvale, CA 94087, USA Absrac Ispred by he Drac oao, a ew se of symbols, he Probably Bracke Noao (PBN) s proposed for probably modelg.

More information

Discrete Mathematics and Probability Theory Fall 2016 Seshia and Walrand DIS 10b

Discrete Mathematics and Probability Theory Fall 2016 Seshia and Walrand DIS 10b CS 70 Dscrete Mathematcs ad Probablty Theory Fall 206 Sesha ad Walrad DIS 0b. Wll I Get My Package? Seaky delvery guy of some compay s out delverg packages to customers. Not oly does he had a radom package

More information

Fully Fuzzy Linear Systems Solving Using MOLP

Fully Fuzzy Linear Systems Solving Using MOLP World Appled Sceces Joural 12 (12): 2268-2273, 2011 ISSN 1818-4952 IDOSI Publcaos, 2011 Fully Fuzzy Lear Sysems Solvg Usg MOLP Tofgh Allahvraloo ad Nasser Mkaelvad Deparme of Mahemacs, Islamc Azad Uversy,

More information

For the plane motion of a rigid body, an additional equation is needed to specify the state of rotation of the body.

For the plane motion of a rigid body, an additional equation is needed to specify the state of rotation of the body. The kecs of rgd bodes reas he relaoshps bewee he exeral forces acg o a body ad he correspodg raslaoal ad roaoal moos of he body. he kecs of he parcle, we foud ha wo force equaos of moo were requred o defe

More information

Solving fuzzy linear programming problems with piecewise linear membership functions by the determination of a crisp maximizing decision

Solving fuzzy linear programming problems with piecewise linear membership functions by the determination of a crisp maximizing decision Frs Jo Cogress o Fuzzy ad Iellge Sysems Ferdows Uversy of Mashhad Ira 9-3 Aug 7 Iellge Sysems Scefc Socey of Ira Solvg fuzzy lear programmg problems wh pecewse lear membershp fucos by he deermao of a crsp

More information

Lecture Slides for INTRODUCTION TO. Machine Learning. ETHEM ALPAYDIN The MIT Press,

Lecture Slides for INTRODUCTION TO. Machine Learning. ETHEM ALPAYDIN The MIT Press, Lecure Sldes for INTRDUCTIN T Machne Learnng ETHEM ALAYDIN The MIT ress, 2004 alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/2ml CHATER 3: Hdden Marov Models Inroducon Modelng dependences n npu; no

More information

4. THE DENSITY MATRIX

4. THE DENSITY MATRIX 4. THE DENSTY MATRX The desy marx or desy operaor s a alerae represeao of he sae of a quaum sysem for whch we have prevously used he wavefuco. Alhough descrbg a quaum sysem wh he desy marx s equvale o

More information

Complementary Tree Paired Domination in Graphs

Complementary Tree Paired Domination in Graphs IOSR Joural of Mahemacs (IOSR-JM) e-issn: 2278-5728, p-issn: 239-765X Volume 2, Issue 6 Ver II (Nov - Dec206), PP 26-3 wwwosrjouralsorg Complemeary Tree Pared Domao Graphs A Meeaksh, J Baskar Babujee 2

More information

RATIO ESTIMATORS USING CHARACTERISTICS OF POISSON DISTRIBUTION WITH APPLICATION TO EARTHQUAKE DATA

RATIO ESTIMATORS USING CHARACTERISTICS OF POISSON DISTRIBUTION WITH APPLICATION TO EARTHQUAKE DATA The 7 h Ieraoal as of Sascs ad Ecoomcs Prague Sepember 9-0 Absrac RATIO ESTIMATORS USING HARATERISTIS OF POISSON ISTRIBUTION WITH APPLIATION TO EARTHQUAKE ATA Gamze Özel Naural pulaos bolog geecs educao

More information

Interval Estimation. Consider a random variable X with a mean of X. Let X be distributed as X X

Interval Estimation. Consider a random variable X with a mean of X. Let X be distributed as X X ECON 37: Ecoomercs Hypohess Tesg Iervl Esmo Wh we hve doe so fr s o udersd how we c ob esmors of ecoomcs reloshp we wsh o sudy. The queso s how comforble re we wh our esmors? We frs exme how o produce

More information

Quantitative Portfolio Theory & Performance Analysis

Quantitative Portfolio Theory & Performance Analysis 550.447 Quaave Porfolo heory & Performace Aalyss Week February 4 203 Coceps. Assgme For February 4 (hs Week) ead: A&L Chaper Iroduco & Chaper (PF Maageme Evrome) Chaper 2 ( Coceps) Seco (Basc eur Calculaos)

More information

Variants of Pegasos. December 11, 2009

Variants of Pegasos. December 11, 2009 Inroducon Varans of Pegasos SooWoong Ryu bshboy@sanford.edu December, 009 Youngsoo Cho yc344@sanford.edu Developng a new SVM algorhm s ongong research opc. Among many exng SVM algorhms, we wll focus on

More information

Chapter 8. Simple Linear Regression

Chapter 8. Simple Linear Regression Chaper 8. Smple Lear Regresso Regresso aalyss: regresso aalyss s a sascal mehodology o esmae he relaoshp of a respose varable o a se of predcor varable. whe here s jus oe predcor varable, we wll use smple

More information

Clustering (Bishop ch 9)

Clustering (Bishop ch 9) Cluserng (Bshop ch 9) Reference: Daa Mnng by Margare Dunham (a slde source) 1 Cluserng Cluserng s unsupervsed learnng, here are no class labels Wan o fnd groups of smlar nsances Ofen use a dsance measure

More information

Modeling and Predicting Sequences: HMM and (may be) CRF. Amr Ahmed Feb 25

Modeling and Predicting Sequences: HMM and (may be) CRF. Amr Ahmed Feb 25 Modelg d redcg Sequeces: HMM d m be CRF Amr Ahmed 070 Feb 25 Bg cure redcg Sgle Lbel Ipu : A se of feures: - Bg of words docume - Oupu : Clss lbel - Topc of he docume - redcg Sequece of Lbels Noo Noe:

More information

Big O Notation for Time Complexity of Algorithms

Big O Notation for Time Complexity of Algorithms BRONX COMMUNITY COLLEGE of he Ciy Uiversiy of New York DEPARTMENT OF MATHEMATICS AND COMPUTER SCIENCE CSI 33 Secio E01 Hadou 1 Fall 2014 Sepember 3, 2014 Big O Noaio for Time Complexiy of Algorihms Time

More information

( ) ( ) Weibull Distribution: k ti. u u. Suppose t 1, t 2, t n are times to failure of a group of n mechanisms. The likelihood function is

( ) ( ) Weibull Distribution: k ti. u u. Suppose t 1, t 2, t n are times to failure of a group of n mechanisms. The likelihood function is Webll Dsbo: Des Bce Dep of Mechacal & Idsal Egeeg The Uvesy of Iowa pdf: f () exp Sppose, 2, ae mes o fale of a gop of mechasms. The lelhood fco s L ( ;, ) exp exp MLE: Webll 3//2002 page MLE: Webll 3//2002

More information

The Optimal Combination Forecasting Based on ARIMA,VAR and SSM

The Optimal Combination Forecasting Based on ARIMA,VAR and SSM Advaces Compuer, Sgals ad Sysems (206) : 3-7 Clausus Scefc Press, Caada The Opmal Combao Forecasg Based o ARIMA,VAR ad SSM Bebe Che,a, Mgya Jag,b* School of Iformao Scece ad Egeerg, Shadog Uversy, Ja,

More information

Queuing Theory: Memory Buffer Limits on Superscalar Processing

Queuing Theory: Memory Buffer Limits on Superscalar Processing Cle/ Model of I/O Queug Theory: Memory Buffer Lms o Superscalar Processg Cle reques respose Devce Fas CPU s cle for slower I/O servces Buffer sores cle requess ad s a slower server respose rae Laecy Tme

More information

EDUCATION COMMITTEE OF THE SOCIETY OF ACTUARIES ADVANCED TOPICS IN GENERAL INSURANCE STUDY NOTE CREDIBILITY WITH SHIFTING RISK PARAMETERS

EDUCATION COMMITTEE OF THE SOCIETY OF ACTUARIES ADVANCED TOPICS IN GENERAL INSURANCE STUDY NOTE CREDIBILITY WITH SHIFTING RISK PARAMETERS EDUCATION COMMITTEE OF THE SOCIETY OF ACTUARIES ADVANCED TOPICS IN GENERAL INSURANCE STUDY NOTE CREDIBILITY WITH SHIFTING RISK PARAMETERS Suar Klugma, FSA, CERA, PhD Copyrgh 04 Socey of Acuares The Educao

More information

Three Main Questions on HMMs

Three Main Questions on HMMs Mache Learg 0-70/5-78 78 Srg 00 Hdde Marov Model II Erc Xg Lecure Februar 4 00 Readg: Cha. 3 CB Three Ma Quesos o HMMs. Evaluao GIVEN a HMM M ad a sequece FIND Prob M ALGO. Forward. Decodg GIVEN a HMM

More information

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period.

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period. ublc Affars 974 Meze D. Ch Fall Socal Sceces 748 Uversy of Wscos-Madso Sock rces, News ad he Effce Markes Hypohess (rev d //) The rese Value Model Approach o Asse rcg The exbook expresses he sock prce

More information

Deterioration-based Maintenance Management Algorithm

Deterioration-based Maintenance Management Algorithm Aca Polyechca Hugarca Vol. 4 No. 2007 Deerorao-baed Maeace Maageme Algorhm Koréla Ambru-Somogy Iue of Meda Techology Budape Tech Doberdó ú 6 H-034 Budape Hugary a_omogy.korela@rkk.bmf.hu Abrac: The Road

More information

Survival Prediction Based on Compound Covariate under Cox Proportional Hazard Models

Survival Prediction Based on Compound Covariate under Cox Proportional Hazard Models Ieraoal Bomerc Coferece 22/8/3, Kobe JAPAN Survval Predco Based o Compoud Covarae uder Co Proporoal Hazard Models PLoS ONE 7. do:.37/oural.poe.47627. hp://d.plos.org/.37/oural.poe.47627 Takesh Emura Graduae

More information

Extremal graph theory II: K t and K t,t

Extremal graph theory II: K t and K t,t Exremal graph heory II: K ad K, Lecure Graph Theory 06 EPFL Frak de Zeeuw I his lecure, we geeralize he wo mai heorems from he las lecure, from riagles K 3 o complee graphs K, ad from squares K, o complee

More information

Exam Supply Chain Management January 17, 2008

Exam Supply Chain Management January 17, 2008 Exam Supply Cha Maageme Jauary 7, 008 IMPORTANT GUIELINES: The exam s closed book. Sudes may use a calculaor. The formularum s aached a he back of he assgme budle. Please wre your aswers o he blak pages

More information

Mixed Integral Equation of Contact Problem in Position and Time

Mixed Integral Equation of Contact Problem in Position and Time Ieraoal Joural of Basc & Appled Sceces IJBAS-IJENS Vol: No: 3 ed Iegral Equao of Coac Problem Poso ad me. A. Abdou S. J. oaquel Deparme of ahemacs Faculy of Educao Aleadra Uversy Egyp Deparme of ahemacs

More information

Nature and Science, 5(1), 2007, Han and Xu, Multi-variable Grey Model based on Genetic Algorithm and its Application in Urban Water Consumption

Nature and Science, 5(1), 2007, Han and Xu, Multi-variable Grey Model based on Genetic Algorithm and its Application in Urban Water Consumption Naure ad Scece, 5, 7, Ha ad u, ul-varable Grey odel based o Geec Algorhm ad s Applcao Urba Waer Cosumpo ul-varable Grey odel based o Geec Algorhm ad s Applcao Urba Waer Cosumpo Ha Ya*, u Shguo School of

More information

Dynamic Voltage and Frequency Scaling Under a Precise Energy Model Considering Variable and Fixed Components of the System Power Dissipation

Dynamic Voltage and Frequency Scaling Under a Precise Energy Model Considering Variable and Fixed Components of the System Power Dissipation Dyamc Volage ad requecy Scalg Uder a recse Eergy Model Cosderg Varable ad xed Compoes of he Sysem ower Dsspao ABSRAC hs paper preses a dyamc volage ad frequecy scalg (DVS) echque ha mmzes he oal sysem

More information

Moments of Order Statistics from Nonidentically Distributed Three Parameters Beta typei and Erlang Truncated Exponential Variables

Moments of Order Statistics from Nonidentically Distributed Three Parameters Beta typei and Erlang Truncated Exponential Variables Joural of Mahemacs ad Sascs 6 (4): 442-448, 200 SSN 549-3644 200 Scece Publcaos Momes of Order Sascs from Nodecally Dsrbued Three Parameers Bea ype ad Erlag Trucaed Expoeal Varables A.A. Jamoom ad Z.A.

More information

Lecture 6: Learning for Control (Generalised Linear Regression)

Lecture 6: Learning for Control (Generalised Linear Regression) Lecure 6: Learnng for Conrol (Generalsed Lnear Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure 6: RLSC - Prof. Sehu Vjayakumar Lnear Regresson

More information

THE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE

THE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE THE ROYAL STATISTICAL SOCIETY 00 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE PAPER I STATISTICAL THEORY The Socety provdes these solutos to assst caddates preparg for the examatos future years ad for the

More information

USING INPUT PROCESS INDICATORS FOR DYNAMIC DECISION MAKING

USING INPUT PROCESS INDICATORS FOR DYNAMIC DECISION MAKING Proceedgs of he 999 Wer Smulao Coferece P. A. Farrgo, H. B. Nembhard, D. T. Surrock, ad G. W. Evas, eds. USING INPUT PROCESS INDICATORS FOR DYNAMIC DECISION MAKING Mchael Fremer School of Operaos Research

More information

SYRIAN SEISMIC CODE :

SYRIAN SEISMIC CODE : SYRIAN SEISMIC CODE 2004 : Two sac mehods have bee ssued Syra buldg code 2004 o calculae he laeral sesmc forces he buldg. The Frs Sac Mehod: I s he same mehod he prevous code (995) wh few modfcaos. I s

More information

1 n. w = How much information can the network store? 4. RECURRENT NETWORKS. One stored pattern, x (1) : Since the elements of the vector x (1) should

1 n. w = How much information can the network store? 4. RECURRENT NETWORKS. One stored pattern, x (1) : Since the elements of the vector x (1) should 4. RECURRENT NETWORKS Ho much formao ca he eor sore? Ths chaper preses some ypes of recurre eural eors: Frs, eors h (srog egh cosras are reaed, eemplfed by Hopfeld eors, folloed by a se of recurre eors

More information

Probability Bracket Notation, Probability Vectors, Markov Chains and Stochastic Processes. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA, USA

Probability Bracket Notation, Probability Vectors, Markov Chains and Stochastic Processes. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA, USA Probably Bracke Noao, Probably Vecors, Markov Chas ad Sochasc Processes Xg M. Wag Sherma Vsual Lab, Suyvale, CA, USA Table of Coes Absrac page1 1. Iroduco page. PBN ad Tme-depede Dscree Radom Varable.1.

More information

Square law expression is non linear between I D and V GS. Need to operate in appropriate region for linear behaviour. W L

Square law expression is non linear between I D and V GS. Need to operate in appropriate region for linear behaviour. W L MOS Feld-Effec Trassrs (MOSFETs ecure # 4 MOSFET as a Amplfer k ( S Square law express s lear bewee ad. Need perae apprprae reg fr lear behaur. Cpyrgh 004 by Oxfrd Uersy Press, c. MOSFET as a Amplfer S

More information