RADIAL BASIS FUNCTION PROCESS NEURAL NETWORK TRAINING BASED ON GENERALIZED FRÉCHET DISTANCE AND GA-SA HYBRID STRATEGY
|
|
- Michael Townsend
- 6 years ago
- Views:
Transcription
1 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 RADIAL BASIS FUNCTION PROCESS NEURAL NETWORK TRAINING BASED ON GENERALIZED FRÉCHET DISTANCE AND GA-SA HYBRID STRATEGY Wang Bng, Meng Yao-hua, Yu Xao-hong 3 School of Compuer & Informaon Technology, Norheas Peroleum Unversy, Daqng, Chna Communcaon Research Cener, Harbn Insue of Technology, Harbn, Chna 3 Exploraon and Developmen Research Insue, Daqng Olfeld Company, Daqng, Chna ABSTRACT For learnng problem of Radal Bass Funcon Process Neural Newor (RBF-PNN), an opmzaon ranng mehod based on GA combned wh SA s proposed n hs paper. Through buldng generalzed Fréche dsance o measure smlary beween me-varyng funcon samples, he learnng problem of radal bass cenre funcons and connecon weghs s convered no he ranng on correspondng dscree sequence coeffcens. Newor ranng obecve funcon s consruced accordng o he leas square error creron, and global opmzaon solvng of newor parameers s mplemened n feasble soluon space by use of global opmzaon feaure of GA and probablsc umpng propery of SA. The expermen resuls llusrae ha he ranng algorhm mproves he newor ranng effcency and sably. KEYWORDS Radal Bass Funcon Process Neural Newor, Tranng Algorhm, Generalzed Fréche Dsance, GA-SA Hybrd Opmzaon. INTRODUCTION Radal bass funcon neural newor (RBFNN) s a hree-layer feedforward neural newor model and was pu forward by Powell, Broomhead and Lowe[,] n he md-980s. I mplemens nonlnear mappng by change propery parameers of a neuron and mproves learnng rae by lnearzaon of connecon wegh adusmen. RBFNN has been appled n many felds such as funcon approxmaon, splne nerpolaon and paern recognon [3,4]. However, n praccal engneerng, many sysems npus are dependen on me; he npus of exsng RBFNN model are usually consan. They have nohng o do wh he me, ha s, he npu/oupu s nsananeous relaonshp beween geomerc pons. Is nformaon processng mechansm canno drecly reflec he characerscs of me-varyng process npu sgnals and he dynamc effec relaonshp beween varables. Therefore, a radal bass funcon process neural newor (RBF-PNN) s esablshed n leraure [5], and s npus can be me-varyng process sgnal DOI : 0.5/cse
2 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 drecly. Through space-me aggregaon of ernel funcon neurons o fnd sample paern process characerscs, he RBF-PNN has been appled n dynamc sample cluserng and mevaryng sgnal denfcaon effecvely. In pracce, me-varyng sample daa s usually dscree me seres obaned by samplng, he exsng RBF-PNN s unable o deal wh dscree npu drecly. I need o f dscree samples, hus fng error exss. In subsequen orhogonal bass expanson, he fne erms of bass funcon also brng runcaon error. In addon, learnng effcency of radonal RBF-PNN algorhm based on graden descen combned wh bass expanson s low, and a he same me affecs sably and generalzaon ably of he newor ha here are greaer freedom consrans among newor parameers. In recen years, evoluonary compuaon-based opmzaon echnology s wdely used o solve opmal soluon n varous engneerng problems, and has been effecvely appled n complex funcon calculaon, process opmzaon, sysem denfcaon and conrol, and so on [6-8]. Genec algorhm (GA) [9] s a bonc global opmzaon mehod ha smulaes Darwn s naural selecon and naural bologcal evoluon process. I s smple, general, robus, suable for parallel processng, as well as effcen and praccal, and has been wdely appled n varous felds. However, GA has some shorcomngs, e.g. mmaure convergence and slow erave rae. If probablsc umpng propery of smulaed annealng (SA) s made use of, he GA combned wh SA are appled no he ranng of RBF-PNN, hs wll mprove RBF-PNN learnng propery grealy. Fréche dsance [0] s a udgmen o measure he smlary beween polygonal curves. In applcaons, dynamc samples normally conss of a sequence of dscree samplng pons, herefore, Eer and Mannla [] pu forward dscree Fréche dsance on he bass of connuous Fréche dsance, and acheved good applcaon effec n proen srucure predcon[], onlne sgnaure verfcaon [3], ec. Tme-varyng process sgnal can be seen as a onedmensonal curve abou he me; herefore, dscree Fréche dsance can be exended o mevaryng funcon space o measure naure dfference beween npu samples of RBF-PNN. For he RBF-PNN ranng problem, we propose a ranng mehod based on generalzed dscree Fréche dsance combned wh GA-SA opmzaon n hs paper. By consrucng a generalzed Fréche dsance norm o measure he dsance beween dynamc samples n me-varyng funcon space, hen he learnng problem of wegh funcons s ransformed no dscree coeffcen ranng. Accordng o he leas square error creron, he newor ranng obecve funcon s consruced, and GA-SA algorhm s adoped for global opmzaon solvng. I s appled no EEG eye sae recognon and acheves good resuls. RBF-PNN opmzaon ranng seps are also gven n he paper.. DISTANCE MEASUREMENT BETWEEN TIME-VARYING SAMPLES BASED ON GENERALIZED FRÉCHET DISTANCE.. Dscree Fréche Dsance Dscree Fréche dsance s defned as follows: (a) Gven a polygonal chan P =< p, p,, pn > wh n verces, a -wal along P parons he pah no dson non-empy subses { P } =,, such ha P,, =< pn p + n > and = n < n < < n = n. 0 0
3 x ( ) Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 (b) Gven wo polygonal chans A =< a, a,, am > and B =< b, b,, bn >, a pared wal along A and B s a -wal { A } =,, along A and a -wal { B } =,, along B for, eher A = or B = (namely, A or B conans exacly one verex). (c) The cos of a pared wal W = {( A,)} B along A and B s W d ( A,) B max = max(,) ds a b () F ( a,) b A B The dscree Fréche dsance beween wo polygonal chans A and B s d ( A,) B mn( = d,) W A B () The pared wal s also called he Fréche algnmen of A and B. F.. Generalzed Fréche Dsance Beween Tme-Varyng Samples W The dscree Fréche dsance can effecvely measure he dfference beween me-varyng funcons (whch can be regarded as he polygonal chans), whle he npu of he RBF-PNN s a vecor composed of me-varyng funcons. Thus, usng he propery ha Eucldean dsance can mplemen pon arge machng, combnng he dscree Fréche dsance and Eucldean dsance, a generalzed Fréche dsance s esablshed, whch can measure he dsance beween mevaryng funcon vecor samples. Gven wo me-varyng funcon samples X()((),(),...,()) = x x xn and X ()((),(),...,()) = x x x n, her generalzed Fréche dsance d((),()) X X s defned as F n = d((),())(((),())) X X = df x x (3) where df ((),()) x x denoes he dscree Fréche dsance beween dscree samplng pons correspondng o x () and x (). 3. RBF-PNN MODEL 3.. Radal Bass Funcon Process Neuron The srucure and nformaon ransfer flow of a radal bass funcon process neuron are shown n Fgure. x () x () x () n C() () ds K ( ) y 3
4 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 Fgure. Radal bass funcon process neuron In Fgure, X()((),(),...,()) = x x xn s an npu funcon vecor; C()((),(),...,()) = c c cn s a cenre funcon vecor of a radal bass funcon neuron; ds() s a dsance measuremen funcon beween he npu funcon vecor and he cenre funcon vecor, and here we adop he generalzed Fréche dsance defned n Eq.(3). K( ) s a ernel funcon of a radal bass process neuron. I could be a Gauss funcon, a mulquadrc funcon and a hn-plae splne funcon ec. The npu/oupu relaonshp of a radal bass funcon process neuron s defned as: y = K((),()) ds X C (4) 3.. Radal Bass Funcon Process Neural Newor Model A RBF-PNN model s composed of radal bass funcon process neurons and oher ypes of neurons accordng o ceran srucure relaons and nformaon ransmsson process. For convenence, consder a mulple npu/sngle oupu feedforward newor model only wh one radal bass process hdden layer defned by Eq. (4). Is srucure s shown n Fgure. x () C() (),() ds K x (). x () n C() (),() ds K. Cm() (),() ds K y Fgure. RBF-PNN model In Fgure, he npu layer has n neurons, he radal bass process hdden layer has m neurons, and ( =,,,) m are he weghed coeffcens o be adused. The npu/oupu relaonshp of RBF-PNN s: m y = K((),()) ds X C (5) = 3.3. Newor Obecve Funcon Gven K learnng sample funcons ((),) X d where X ()(),(),...,() = x x xn, =,,..., K. Suppose ha y s a real oupu correspondng o he h sample npu. Accordng o he leas square error creron, he newor obecve funcon s defned as below: K ( ) (6) = E = y d 4
5 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 The ernel funcon of he radal bass process neuron adops Gauss funcon: K() d = exp() d (7) where s he wdh of Gauss funcon, and needs o be deermned va newor ranng, hen we have K E = ( y d = ) ds X (),() C = d K m ( exp() = = ) (8) In Eq.(8), snce ds X (),() C s based on generalzed dscree Fréche dsance, f dscree samplng pon number of RBF-PNN npu s S, hen { } s= c s S s he dscree pon sequence correspondng o c (), =,,, n, whch s he h componen of he h radal bass cenre funcon vecor. In concluson, he learnng obecve funcon E of RBF-PNN s a s funcon abou varable parameers { },, S c. s= 4. RBF-PNN TRAINING BASED ON GA-SA ALGORITHM In GA [9], selecon, crossover and muaon consue s man operaons. Parameer encodng, nal populaon, fness funcon choce, genec operaon desgn, conrol parameer seng form he core conen of GA. The obvous advanage of GA s o search many pons n search space a he same me, and s opmal soluon searchng process s nsrucve so as o avod he dmenson dsaser problem exsng n some oher opmzaon algorhms. Smulaed annealng (SA)[4] algorhm s a heursc random search algorhm. Under a ceran nal emperaure, accompaned by emperaure fall, by use of probablsc umpng propery, he global opmal soluon of he obecve funcon s obaned n soluon space. In vew of he respecve feaures of GA and SA, consderng ha SA can effecvely preven GA from rappng n local opmal soluon, GA and SA wll be combned for RBF-PNN ranng o oban he opmal soluon of he problem. The ranng problem of RBF-PNN s o adus parameers o mnmze he obecve funcon s S { } s= E=( E c,,). The concree RBF-PNN ranng seps based on GA-SA s descrbed below: Sep Gven nal smulaed annealng emperaure, and se =. Sep Deermne he populaon sze N, and randomly generae nal populaon. Decmal number s used for chromosome encodng, and he gene number on each chromosome s he number of varables o be opmzed. Se he maxmum erave number M and error precson. Sep 3 Evaluae chromosomes n he populaon. Snce RBF-PNN ranng s he mnmum opmzaon problem for he obecve funcon, he fness funcon s aen as 5
6 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 f = + E (9) Sep 4 Perform genec operaons for chromosomes n he populaon. () Selecon operaon. Adop proporonal selecon operaor, and he probably ha he chromosome X wh fness value f s seleced no he nex generaon s M p = f f (0) = () Crossover operaon. For decmal number encodng, choose arhmec crossover operaor. Chromosomes X and X n he paren perform crossover operaon wh crossover probably p c, he offsprng chromosomes generaed are X = ax + () a X () ' where a s a parameer and a (0,). X = () a X + ax () ' (3) Muaon operaon. Use unform muaon operaor. Each gene n a chromosome X performs muaon operaon wh muaon probably p m, namely unformly dsrbued random number whn he scope of gene value s used o nsead of he orgnal value wh probably p m. Sep 5 Inroduce he opmal reenon sraegy. Sep 6 Each chromosome n he populaon performs smulaed annealng operaon. () Creae new gene g ' ()() = g + by use of SA sae generaon funcon where ( mn, max) are he random dsurbance and (mn,max) are he scope of correspondng 0 0 gene value. () Calculae obecve funcon dfference value C beween (3) Calculae accepance probably p = mn[,exp( C /)]. r g ' () and g(). ' (4) If pr > random[0,] hen g()() = g ; oherwse g() remans unchanged. (5) Inroduce he opmal reenon sraegy. (6) Anneal emperaure by usng he annealng emperaure funcon + = v where v (0,) s he annealng emperaure rae, and se = +. Sep 7 If he obecve funcon value correspondng o he opmal chromosome mees error requremens or he eraon process acheves he maxmum number M, urn o Sep 8; oherwse reurn o Sep 3. Sep 8 Decode he opmal chromosome searched by GA and oban he opmal parameers of RBF-PNN. 5. APPLICATIONS IN EEG EYE STATE CLASSIFICATION 6
7 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 The smulaon expermens use EEG Eye Sae Daa Se from UCI. The daa se was colleced for examnng a classfer for sequenal me seres. The daa se consss of 4 EEG values and a value ndcang he eye sae, and '' ndcaes he eye-closed and '0' he eye-open sae. All daa s from one connuous EEG measuremen wh he Emov EEG Neuroheadse. The par of samples wh he eye-closed sae and he eye-open sae are shown n Fgure (a) Fgure 3. (a) The eye-closed sae samples (b) The eye-open sae samples In he expermens, he ranng se ncludes 60 samples wh each 30 samples separaely from wo saes, and he es se consss of 60 samples. The RBF-PNN srucure parameers are : mevaryng npu node n =, 8 RBF hdden nodes m = 8 and common oupu node. The dscree samplng number s S = 4, he populaon sze s 5,opmzaon eraon number s 000, and he error precson s As he comparson, we also use graden descen (GD) algorhm [5] and GA o ran he RBF-PNN. Each algorhm runs 0 mes, and hen hree algorhms are separaely used o denfy he es samples. The average denfcaon accuracy s shown n Fgure 4. I can be seen ha he SA s on no GA prevens no local mnmum effecvely and mproves he denfcaon resul of he newors. (b) 00 Accuracy (%) GD GA GA-SA 7
8 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 Fgure 4. Accuracy of hree denfcaon algorhms In addon, he ranng algorhms based on generalzed Fréche dsance and on orhogonal bass expanson are respecvely adoped o ran he RBF-PNN. The accuracy based on generalzed Fréche dsance s superor o on orhogonal bass expanson. Ths s because he algorhm based on orhogonal bass expanson has fng error and runcaon error nevably. 6. CONCLUSIONS A RBF-PNN ranng mehod based on GA and SA hybrd opmzaon sraegy s pu forward n hs paper. Accordng o he newor ranng obecve funcon, by buldng a generalzed Fréche dsance used o measure smlary beween me-varyng funcon samples, he funconal opmzaon problem of RBF-PNN ranng s convered no exremal opmzaon solvng of mulvarae funcon, and he global opmzaon soluon of he newor parameers s obaned n he feasble soluon space. I also has ceran reference value for oher machne learnng and complex funcon opmzaon problems. REFERENCES [] Powel M.J. D. (985) Radcal bass funcons for mulvarable nerpolaon: A revew, IMA conference on Algorhms for he Approxmaon of Funcons and Daa, RMCS, Shrvenham, UK: pp [] Broomhead D. S, Lowe D. (988) Mulvarab le funcon nerpolaon and adapve newors, Complex Sysems, No., pp [3] He S, Zou Y, Quan D, e al. (0) Applcaon of RBF Neural Newor and ANFIS on he Predcon of Corroson Rae of Ppelne Seel n Sol, Recen Advances n Compuer Scence and Informaon Engneerng. Sprnger Berln Hedelberg, pp [4] Kang L, Jan-Xun Peng e al. (009) Two -Sage Mxed Dscree Connuous Idenfcaon of Radal Bass Funcon (RBF) Neural Models for Nonlnear Sysems, IEEE Transacons on Crcus and Sysems, Vol. 56, No.3, pp [5] Xu Shao-hua, He Xn-gu. (004) Research and applcaons of radal bass process neural newors,journal of Beng Unversy of Aeronaucs and Asronaucs, Vol. 30, No., pp4-7. [6] De Jong K. (0) Evoluonary compuaon: a unfed approach[c] Proceedngs of he foureenh nernaonal conference on Genec and evoluonary compuaon conference companon. ACM, pp [7] Smon D. (0) A probablsc analyss of a s mplfed bogeography-based opmzaon algorhm, Evoluonary compuaon, Vol. 9, No., pp [8] Vdal T, Cranc T G, Gendreau M, e al. (0) A hybrd genec algorhm for muldepo and perodc vehcle roung problems, Operaons Research, Vol. 60, No. 3, pp [9] De Govann L, Pezzella F. (00) An mproved genec algorhm for he dsrbued and flexble ob-shop schedulng problem, European ournal of operaonal research, 00, 00(): [0] Al H, Godau M, (99) Measurng he Resemblance of Polygonal Curves, In Proceedngs of he 8h Annual Symposum on Compuaonal Geomery, pp [] Eer T, Mannla H, (994) Compung Dscree Fréche Dsance, Vena: Informaon Sysems Deparmen, Techncal Unversy of Vena. [] Mnghu Jang, Yng Xu, Bnha Zhu, (007) Proen Srucure -srucure Algnmen wh Dscree Fréche Dsance, In Proceedngs of he 5h Asa-Pacfc Bonformacs Conference, APBC, pp3-4. [3] Janbn Zheng, Xaole Gao, Enq Zhan, Zhangcan Huang, (008) Algorhm of On-lne Handwrng Sgnaure Verfcaon Based on Dscree Fréche Dsance, In Proceedngs of he 3rd Inernaonal Symposum on Inellgence Compuaon and Applcaons, ISICA 008, pp
9 Compuer Scence & Engneerng: An Inernaonal Journal (CSEIJ), Vol. 3, No. 6, December 03 [4] Avnaash M R, Kumar G R, Bhargav K A, e al. (03) Smulaed annealng approach o soluon of mul-obecve opmal economc dspach, Inellgen Sysems and Conrol (ISCO), 03 7h Inernaonal Conference on. IEEE, pp 7-3. Auhors Wang Bng, born n 98, s a Ph.D. canddae n Norheas Peroleum Unversy. She receved her Bachelor and Maser degrees n Norheas Peroleum Unversy, DaQng provnce Chna n 7/004, 4/007 respecvely. From 007 o now she s a lecurer a Deparmen of Compuer & Informaon Technology, Norheas Peroleum Unversy. She maors n process neural newors, nellgence nformaon processng and daa mnng ec. She has publshed more han 5 refereed ournal and conference papers. Meng Yao-hua, born n 98, s a Ph.D. canddae n he Harbn Insue of Technology, Communcaon Research Cener. He receved hs Bachelor and Maser degrees n Norheas Peroleum Unversy, DaQng provnce Chna n 7/003, 8/007 respecvely. From 007 o 00 he was a lecurer a Deparmen of Elecrcal and Elecronc Engneerng, Helongang Bay Agrculural Unversy. Durng he year 0, he was a vsng scholar a Harbn Insue of Technology. In Augus 0, he on Cenre for massve daase mnng a Harbn Insue of Technology as a researcher. Hs research of neress ncludes process neural newors, sgnal and nformaon processng ec. Yu Xao-hong, born n 98, s an engneer n Exploraon and Developmen Research Insue of Daqng Olfeld Company, Chna. She receved her Bachelor and Maser degrees n Norheas Peroleum Unversy, DaQng provnce Chna n 7/005, 4/008 respecvely. From 008 o now she wors n Daqng Olfeld Company. Her neress nclude ol feld daa denfcaon, arfcal neural newors, nellgen nformaon processng ec. 9
The Analysis of the Thickness-predictive Model Based on the SVM Xiu-ming Zhao1,a,Yan Wang2,band Zhimin Bi3,c
h Naonal Conference on Elecrcal, Elecroncs and Compuer Engneerng (NCEECE The Analyss of he Thcknesspredcve Model Based on he SVM Xumng Zhao,a,Yan Wang,band Zhmn B,c School of Conrol Scence and Engneerng,
More informationOutline. 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 informationParameter Estimation of Three-Phase Induction Motor by Using Genetic Algorithm
360 Journal of Elecrcal Engneerng & Technology Vol. 4, o. 3, pp. 360~364, 009 Parameer Esmaon of Three-Phase Inducon Moor by Usng Genec Algorhm Seesa Jangj and Panhep Laohacha* Absrac Ths paper suggess
More informationGenetic Algorithm in Parameter Estimation of Nonlinear Dynamic Systems
Genec Algorhm n Parameer Esmaon of Nonlnear Dynamc Sysems E. Paeraks manos@egnaa.ee.auh.gr V. Perds perds@vergna.eng.auh.gr Ah. ehagas kehagas@egnaa.ee.auh.gr hp://skron.conrol.ee.auh.gr/kehagas/ndex.hm
More informationMANY real-world applications (e.g. production
Barebones Parcle Swarm for Ineger Programmng Problems Mahamed G. H. Omran, Andres Engelbrech and Ayed Salman Absrac The performance of wo recen varans of Parcle Swarm Opmzaon (PSO) when appled o Ineger
More informationShort-Term Load Forecasting Using PSO-Based Phase Space Neural Networks
Proceedngs of he 5h WSEAS In. Conf. on SIMULATION, MODELING AND OPTIMIZATION, Corfu, Greece, Augus 7-9, 005 (pp78-83) Shor-Term Load Forecasng Usng PSO-Based Phase Space Neural Neworks Jang Chuanwen, Fang
More informationAttribute Reduction Algorithm Based on Discernibility Matrix with Algebraic Method GAO Jing1,a, Ma Hui1, Han Zhidong2,b
Inernaonal Indusral Informacs and Compuer Engneerng Conference (IIICEC 05) Arbue educon Algorhm Based on Dscernbly Marx wh Algebrac Mehod GAO Jng,a, Ma Hu, Han Zhdong,b Informaon School, Capal Unversy
More informationOn computing differential transform of nonlinear non-autonomous functions and its applications
On compung dfferenal ransform of nonlnear non-auonomous funcons and s applcaons Essam. R. El-Zahar, and Abdelhalm Ebad Deparmen of Mahemacs, Faculy of Scences and Humanes, Prnce Saam Bn Abdulazz Unversy,
More informationTheoretical Analysis of Biogeography Based Optimization Aijun ZHU1,2,3 a, Cong HU1,3, Chuanpei XU1,3, Zhi Li1,3
6h Inernaonal Conference on Machnery, Maerals, Envronmen, Boechnology and Compuer (MMEBC 6) Theorecal Analyss of Bogeography Based Opmzaon Aun ZU,,3 a, Cong U,3, Chuanpe XU,3, Zh L,3 School of Elecronc
More informationIntroduction to Boosting
Inroducon o Boosng Cynha Rudn PACM, Prnceon Unversy Advsors Ingrd Daubeches and Rober Schapre Say you have a daabase of news arcles, +, +, -, -, +, +, -, -, +, +, -, -, +, +, -, + where arcles are labeled
More informationISSN MIT Publications
MIT Inernaonal Journal of Elecrcal and Insrumenaon Engneerng Vol. 1, No. 2, Aug 2011, pp 93-98 93 ISSN 2230-7656 MIT Publcaons A New Approach for Solvng Economc Load Dspach Problem Ansh Ahmad Dep. of Elecrcal
More informationLecture 11 SVM cont
Lecure SVM con. 0 008 Wha we have done so far We have esalshed ha we wan o fnd a lnear decson oundary whose margn s he larges We know how o measure he margn of a lnear decson oundary Tha s: he mnmum geomerc
More informationVariants 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 informationWiH Wei He
Sysem Idenfcaon of onlnear Sae-Space Space Baery odels WH We He wehe@calce.umd.edu Advsor: Dr. Chaochao Chen Deparmen of echancal Engneerng Unversy of aryland, College Par 1 Unversy of aryland Bacground
More informationAn introduction to Support Vector Machine
An nroducon o Suppor Vecor Machne 報告者 : 黃立德 References: Smon Haykn, "Neural Neworks: a comprehensve foundaon, second edon, 999, Chaper 2,6 Nello Chrsann, John Shawe-Tayer, An Inroducon o Suppor Vecor Machnes,
More informationCS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4
CS434a/54a: Paern Recognon Prof. Olga Veksler Lecure 4 Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped
More informationApproximate Analytic Solution of (2+1) - Dimensional Zakharov-Kuznetsov(Zk) Equations Using Homotopy
Arcle Inernaonal Journal of Modern Mahemacal Scences, 4, (): - Inernaonal Journal of Modern Mahemacal Scences Journal homepage: www.modernscenfcpress.com/journals/jmms.aspx ISSN: 66-86X Florda, USA Approxmae
More informationRobust and Accurate Cancer Classification with Gene Expression Profiling
Robus and Accurae Cancer Classfcaon wh Gene Expresson Proflng (Compuaonal ysems Bology, 2005) Auhor: Hafeng L, Keshu Zhang, ao Jang Oulne Background LDA (lnear dscrmnan analyss) and small sample sze problem
More informationFTCS Solution to the Heat Equation
FTCS Soluon o he Hea Equaon ME 448/548 Noes Gerald Reckenwald Porland Sae Unversy Deparmen of Mechancal Engneerng gerry@pdxedu ME 448/548: FTCS Soluon o he Hea Equaon Overvew Use he forward fne d erence
More informationStudy on Distribution Network Reconfiguration with Various DGs
Inernaonal Conference on Maerals Engneerng and Informaon Technology Applcaons (MEITA 205) Sudy on Dsrbuon ework Reconfguraon wh Varous DGs Shengsuo u a, Y Dng b and Zhru Lang c School of Elecrcal Engneerng,
More informationCubic Bezier Homotopy Function for Solving Exponential Equations
Penerb Journal of Advanced Research n Compung and Applcaons ISSN (onlne: 46-97 Vol. 4, No.. Pages -8, 6 omoopy Funcon for Solvng Eponenal Equaons S. S. Raml *,,. Mohamad Nor,a, N. S. Saharzan,b and M.
More informationOn One Analytic Method of. Constructing Program Controls
Appled Mahemacal Scences, Vol. 9, 05, no. 8, 409-407 HIKARI Ld, www.m-hkar.com hp://dx.do.org/0.988/ams.05.54349 On One Analyc Mehod of Consrucng Program Conrols A. N. Kvko, S. V. Chsyakov and Yu. E. Balyna
More informationRobustness Experiments with Two Variance Components
Naonal Insue of Sandards and Technology (NIST) Informaon Technology Laboraory (ITL) Sascal Engneerng Dvson (SED) Robusness Expermens wh Two Varance Componens by Ana Ivelsse Avlés avles@ns.gov Conference
More informationDynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005
Dynamc Team Decson Theory EECS 558 Proec Shruvandana Sharma and Davd Shuman December 0, 005 Oulne Inroducon o Team Decson Theory Decomposon of he Dynamc Team Decson Problem Equvalence of Sac and Dynamc
More informationIntroduction ( Week 1-2) Course introduction A brief introduction to molecular biology A brief introduction to sequence comparison Part I: Algorithms
Course organzaon Inroducon Wee -2) Course nroducon A bref nroducon o molecular bology A bref nroducon o sequence comparson Par I: Algorhms for Sequence Analyss Wee 3-8) Chaper -3, Models and heores» Probably
More informationConstrained-Storage Variable-Branch Neural Tree for. Classification
Consraned-Sorage Varable-Branch Neural Tree for Classfcaon Shueng-Ben Yang Deparmen of Dgal Conen of Applcaon and Managemen Wenzao Ursulne Unversy of Languages 900 Mnsu s oad Kaohsng 807, Tawan. Tel :
More informationNeural Networks-Based Time Series Prediction Using Long and Short Term Dependence in the Learning Process
Neural Neworks-Based Tme Seres Predcon Usng Long and Shor Term Dependence n he Learnng Process J. Puchea, D. Paño and B. Kuchen, Absrac In hs work a feedforward neural neworksbased nonlnear auoregresson
More informationAn Effective TCM-KNN Scheme for High-Speed Network Anomaly Detection
Vol. 24, November,, 200 An Effecve TCM-KNN Scheme for Hgh-Speed Nework Anomaly eecon Yang L Chnese Academy of Scences, Bejng Chna, 00080 lyang@sofware.c.ac.cn Absrac. Nework anomaly deecon has been a ho
More informationParticle Swarm Optimization Algorithm with Reverse-Learning and Local-Learning Behavior
35 JOURNAL OF SOFTWARE, VOL. 9, NO. 2, FEBRUARY 214 Parcle Swarm Opmzaon Algorhm wh Reverse-Learnng and Local-Learnng Behavor Xuewen Xa Naonal Engneerng Research Cener for Saelle Posonng Sysem, Wuhan Unversy,
More informationTSS = SST + SSE An orthogonal partition of the total SS
ANOVA: Topc 4. Orhogonal conrass [ST&D p. 183] H 0 : µ 1 = µ =... = µ H 1 : The mean of a leas one reamen group s dfferen To es hs hypohess, a basc ANOVA allocaes he varaon among reamen means (SST) equally
More informationHEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD
Journal of Appled Mahemacs and Compuaonal Mechancs 3, (), 45-5 HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Sansław Kukla, Urszula Sedlecka Insue of Mahemacs,
More informationLecture 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 informationIterative Learning Control and Applications in Rehabilitation
Ierave Learnng Conrol and Applcaons n Rehablaon Yng Tan The Deparmen of Elecrcal and Elecronc Engneerng School of Engneerng The Unversy of Melbourne Oulne 1. A bref nroducon of he Unversy of Melbourne
More informationBayes rule for a classification problem INF Discriminant functions for the normal density. Euclidean distance. Mahalanobis distance
INF 43 3.. Repeon Anne Solberg (anne@f.uo.no Bayes rule for a classfcaon problem Suppose we have J, =,...J classes. s he class label for a pxel, and x s he observed feaure vecor. We can use Bayes rule
More informationPattern Classification (III) & Pattern Verification
Preare by Prof. Hu Jang CSE638 --4 CSE638 3. Seech & Language Processng o.5 Paern Classfcaon III & Paern Verfcaon Prof. Hu Jang Dearmen of Comuer Scence an Engneerng York Unversy Moel Parameer Esmaon Maxmum
More informationTight results for Next Fit and Worst Fit with resource augmentation
Tgh resuls for Nex F and Wors F wh resource augmenaon Joan Boyar Leah Epsen Asaf Levn Asrac I s well known ha he wo smple algorhms for he classc n packng prolem, NF and WF oh have an approxmaon rao of
More informationThis document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore.
Ths documen s downloaded from DR-NTU, Nanyang Technologcal Unversy Lbrary, Sngapore. Tle A smplfed verb machng algorhm for word paron n vsual speech processng( Acceped verson ) Auhor(s) Foo, Say We; Yong,
More informationV.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS
R&RATA # Vol.) 8, March FURTHER AALYSIS OF COFIDECE ITERVALS FOR LARGE CLIET/SERVER COMPUTER ETWORKS Vyacheslav Abramov School of Mahemacal Scences, Monash Unversy, Buldng 8, Level 4, Clayon Campus, Wellngon
More informationComb Filters. Comb Filters
The smple flers dscussed so far are characered eher by a sngle passband and/or a sngle sopband There are applcaons where flers wh mulple passbands and sopbands are requred Thecomb fler s an example of
More informationLinear Response Theory: The connection between QFT and experiments
Phys540.nb 39 3 Lnear Response Theory: The connecon beween QFT and expermens 3.1. Basc conceps and deas Q: ow do we measure he conducvy of a meal? A: we frs nroduce a weak elecrc feld E, and hen measure
More informationEEL 6266 Power System Operation and Control. Chapter 5 Unit Commitment
EEL 6266 Power Sysem Operaon and Conrol Chaper 5 Un Commmen Dynamc programmng chef advanage over enumeraon schemes s he reducon n he dmensonaly of he problem n a src prory order scheme, here are only N
More informationRelative controllability of nonlinear systems with delays in control
Relave conrollably o nonlnear sysems wh delays n conrol Jerzy Klamka Insue o Conrol Engneerng, Slesan Techncal Unversy, 44- Glwce, Poland. phone/ax : 48 32 37227, {jklamka}@a.polsl.glwce.pl Keywor: Conrollably.
More informationVolatility Interpolation
Volaly Inerpolaon Prelmnary Verson March 00 Jesper Andreasen and Bran Huge Danse Mares, Copenhagen wan.daddy@danseban.com brno@danseban.com Elecronc copy avalable a: hp://ssrn.com/absrac=69497 Inro Local
More informationLearning Objectives. Self Organization Map. Hamming Distance(1/5) Introduction. Hamming Distance(3/5) Hamming Distance(2/5) 15/04/2015
/4/ Learnng Objecves Self Organzaon Map Learnng whou Exaples. Inroducon. MAXNET 3. Cluserng 4. Feaure Map. Self-organzng Feaure Map 6. Concluson 38 Inroducon. Learnng whou exaples. Daa are npu o he syse
More informationHongyuan Gao* and Ming Diao
In. J. odellng, Idenfcaon and Conrol, Vol. X, No. Y, 200X Culural frework algorhm and s applcaon for dgal flers desgn Hongyuan Gao* and ng Dao College of Informaon and Communcaon Engneerng, Harbn Engneerng
More informationChapter Lagrangian Interpolation
Chaper 5.4 agrangan Inerpolaon Afer readng hs chaper you should be able o:. dere agrangan mehod of nerpolaon. sole problems usng agrangan mehod of nerpolaon and. use agrangan nerpolans o fnd deraes and
More informationMachine Learning Linear Regression
Machne Learnng Lnear Regresson Lesson 3 Lnear Regresson Bascs of Regresson Leas Squares esmaon Polynomal Regresson Bass funcons Regresson model Regularzed Regresson Sascal Regresson Mamum Lkelhood (ML)
More informationComputing Relevance, Similarity: The Vector Space Model
Compung Relevance, Smlary: The Vecor Space Model Based on Larson and Hears s sldes a UC-Bereley hp://.sms.bereley.edu/courses/s0/f00/ aabase Managemen Sysems, R. Ramarshnan ocumen Vecors v ocumens are
More informationHighway Passenger Traffic Volume Prediction of Cubic Exponential Smoothing Model Based on Grey System Theory
Inernaonal Conference on on Sof Compung n Informaon Communcaon echnology (SCIC 04) Hghway Passenger raffc Volume Predcon of Cubc Exponenal Smoohng Model Based on Grey Sysem heory Wenwen Lu, Yong Qn, Honghu
More informationCS286.2 Lecture 14: Quantum de Finetti Theorems II
CS286.2 Lecure 14: Quanum de Fne Theorems II Scrbe: Mara Okounkova 1 Saemen of he heorem Recall he las saemen of he quanum de Fne heorem from he prevous lecure. Theorem 1 Quanum de Fne). Le ρ Dens C 2
More informationA NEW TECHNIQUE FOR SOLVING THE 1-D BURGERS EQUATION
S19 A NEW TECHNIQUE FOR SOLVING THE 1-D BURGERS EQUATION by Xaojun YANG a,b, Yugu YANG a*, Carlo CATTANI c, and Mngzheng ZHU b a Sae Key Laboraory for Geomechancs and Deep Underground Engneerng, Chna Unversy
More informationSOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β
SARAJEVO JOURNAL OF MATHEMATICS Vol.3 (15) (2007), 137 143 SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β M. A. K. BAIG AND RAYEES AHMAD DAR Absrac. In hs paper, we propose
More informationAdvanced Machine Learning & Perception
Advanced Machne Learnng & Percepon Insrucor: Tony Jebara SVM Feaure & Kernel Selecon SVM Eensons Feaure Selecon (Flerng and Wrappng) SVM Feaure Selecon SVM Kernel Selecon SVM Eensons Classfcaon Feaure/Kernel
More informationJohn Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany
Herarchcal Markov Normal Mxure models wh Applcaons o Fnancal Asse Reurns Appendx: Proofs of Theorems and Condonal Poseror Dsrbuons John Geweke a and Gann Amsano b a Deparmens of Economcs and Sascs, Unversy
More informationTime-interval analysis of β decay. V. Horvat and J. C. Hardy
Tme-nerval analyss of β decay V. Horva and J. C. Hardy Work on he even analyss of β decay [1] connued and resuled n he developmen of a novel mehod of bea-decay me-nerval analyss ha produces hghly accurae
More informationPARTICLE SWARM OPTIMIZATION BASED ON BOTTLENECK MACHINE FOR JOBSHOP SCHEDULING
Proceedng 7 h Inernaonal Semnar on Indusral Engneerng and Managemen PARTICLE SWARM OPTIMIZATION BASED ON BOTTLENECK MACHINE FOR JOBSHOP SCHEDULING Rahm Mauldya Indusral Engneerng Deparmen, Indusral Engneerng
More informationConvergence and Calculation Speed of Genetic Algorithm in Structural Engineering Optimization
Carographca Snca 22(2) p.p.49-54. 9. Kmes D S Sellers P J. (985) Inferrng hemsphercal reflecance of he earh s surface for global energy budges from remoely sensed nadr or dreconal radance values. Remoe
More informationCS 536: Machine Learning. Nonparametric Density Estimation Unsupervised Learning - Clustering
CS 536: Machne Learnng Nonparamerc Densy Esmaon Unsupervsed Learnng - Cluserng Fall 2005 Ahmed Elgammal Dep of Compuer Scence Rugers Unversy CS 536 Densy Esmaon - Cluserng - 1 Oulnes Densy esmaon Nonparamerc
More informationDetection of Waving Hands from Images Using Time Series of Intensity Values
Deecon of Wavng Hands from Images Usng Tme eres of Inensy Values Koa IRIE, Kazunor UMEDA Chuo Unversy, Tokyo, Japan re@sensor.mech.chuo-u.ac.jp, umeda@mech.chuo-u.ac.jp Absrac Ths paper proposes a mehod
More informationA NOVEL NETWORK METHOD DESIGNING MULTIRATE FILTER BANKS AND WAVELETS
A NOVEL NEWORK MEHOD DESIGNING MULIRAE FILER BANKS AND WAVELES Yng an Deparmen of Elecronc Engneerng and Informaon Scence Unversy of Scence and echnology of Chna Hefe 37, P. R. Chna E-mal: yan@usc.edu.cn
More informationExistence and Uniqueness Results for Random Impulsive Integro-Differential Equation
Global Journal of Pure and Appled Mahemacs. ISSN 973-768 Volume 4, Number 6 (8), pp. 89-87 Research Inda Publcaons hp://www.rpublcaon.com Exsence and Unqueness Resuls for Random Impulsve Inegro-Dfferenal
More informationTesting a new idea to solve the P = NP problem with mathematical induction
Tesng a new dea o solve he P = NP problem wh mahemacal nducon Bacground P and NP are wo classes (ses) of languages n Compuer Scence An open problem s wheher P = NP Ths paper ess a new dea o compare he
More informationA Novel Iron Loss Reduction Technique for Distribution Transformers. Based on a Combined Genetic Algorithm - Neural Network Approach
A Novel Iron Loss Reducon Technque for Dsrbuon Transformers Based on a Combned Genec Algorhm - Neural Nework Approach Palvos S. Georglaks Nkolaos D. Doulams Anasasos D. Doulams Nkos D. Hazargyrou and Sefanos
More informationThe Dynamic Programming Models for Inventory Control System with Time-varying Demand
The Dynamc Programmng Models for Invenory Conrol Sysem wh Tme-varyng Demand Truong Hong Trnh (Correspondng auhor) The Unversy of Danang, Unversy of Economcs, Venam Tel: 84-236-352-5459 E-mal: rnh.h@due.edu.vn
More informationLecture VI Regression
Lecure VI Regresson (Lnear Mehods for Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure VI: MLSC - Dr. Sehu Vjayakumar Lnear Regresson Model M
More informationClustering (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 informationHongbin Dong Computer Science and Technology College Harbin Engineering University Harbin, China
Hongbn Dong Compuer Scence and Harbn Engneerng Unversy Harbn, Chna donghongbn@hrbeu.edu.cn Xue Yang Compuer Scence and Harbn Engneerng Unversy Harbn, Chna yangxue@hrbeu.edu.cn Xuyang Teng Compuer Scence
More informationOnline Supplement for Dynamic Multi-Technology. Production-Inventory Problem with Emissions Trading
Onlne Supplemen for Dynamc Mul-Technology Producon-Invenory Problem wh Emssons Tradng by We Zhang Zhongsheng Hua Yu Xa and Baofeng Huo Proof of Lemma For any ( qr ) Θ s easy o verfy ha he lnear programmng
More informationEffective Task Scheduling and Dynamic Resource Optimization based on Heuristic Algorithms in Cloud Computing Environment
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL 11, NO 12, Dec 2017 5780 Copyrgh c2017 KSII Effecve Task Schedulng and Dynamc Resource Opmzaon based on Heursc Algorhms n Cloud Compung Envronmen
More informationPARTICLE SWARM OPTIMIZATION FOR INTERACTIVE FUZZY MULTIOBJECTIVE NONLINEAR PROGRAMMING. T. Matsui, M. Sakawa, K. Kato, T. Uno and K.
Scenae Mahemacae Japoncae Onlne, e-2008, 1 13 1 PARTICLE SWARM OPTIMIZATION FOR INTERACTIVE FUZZY MULTIOBJECTIVE NONLINEAR PROGRAMMING T. Masu, M. Sakawa, K. Kao, T. Uno and K. Tamada Receved February
More informationRefined Binary Particle Swarm Optimization and Application in Power System
Po-Hung Chen, Cheng-Chen Kuo, Fu-Hsen Chen, Cheng-Chuan Chen Refned Bnary Parcle Swarm Opmzaon and Applcaon n Power Sysem PO-HUNG CHEN, CHENG-CHIEN KUO, FU-HSIEN CHEN, CHENG-CHUAN CHEN* Deparmen of Elecrcal
More informationDEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL
DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL Sco Wsdom, John Hershey 2, Jonahan Le Roux 2, and Shnj Waanabe 2 Deparmen o Elecrcal Engneerng, Unversy o Washngon, Seale, WA, USA
More informationSingle-loop System Reliability-Based Design & Topology Optimization (SRBDO/SRBTO): A Matrix-based System Reliability (MSR) Method
10 h US Naonal Congress on Compuaonal Mechancs Columbus, Oho 16-19, 2009 Sngle-loop Sysem Relably-Based Desgn & Topology Opmzaon (SRBDO/SRBTO): A Marx-based Sysem Relably (MSR) Mehod Tam Nguyen, Junho
More informationP R = P 0. The system is shown on the next figure:
TPG460 Reservor Smulaon 08 page of INTRODUCTION TO RESERVOIR SIMULATION Analycal and numercal soluons of smple one-dmensonal, one-phase flow equaons As an nroducon o reservor smulaon, we wll revew he smples
More informationAlgorithm Research on Moving Object Detection of Surveillance Video Sequence *
Opcs and Phooncs Journal 03 3 308-3 do:0.436/opj.03.3b07 Publshed Onlne June 03 (hp://www.scrp.org/journal/opj) Algorhm Research on Movng Objec Deecon of Survellance Vdeo Sequence * Kuhe Yang Zhmng Ca
More informationIntroduction to Compact Dynamical Modeling. III.1 Reducing Linear Time Invariant Systems. Luca Daniel Massachusetts Institute of Technology
SF & IH Inroducon o Compac Dynamcal Modelng III. Reducng Lnear me Invaran Sysems Luca Danel Massachuses Insue of echnology Course Oulne Quck Sneak Prevew I. Assemblng Models from Physcal Problems II. Smulang
More informationTHE PREDICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS
THE PREICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS INTROUCTION The wo dmensonal paral dfferenal equaons of second order can be used for he smulaon of compeve envronmen n busness The arcle presens he
More informationStudy on Intelligent Temperature Controller Based on Expert Controlling
Sensors & Transducers 2014 by IFSA Publshng, S. L. hp://www.sensorsporal.com Sudy on Inellgen Temperaure Conroller Based on Exper Conrollng Jan-Hu MA, Peng GUO, Ka MENG College of Mechancal Engneerng,
More informationOff line signatures Recognition System using Discrete Cosine Transform and VQ/HMM
Ausralan Journal of Basc and Appled Scences, 6(12): 423-428, 2012 ISSN 1991-8178 Off lne sgnaures Recognon Sysem usng Dscree Cosne Transform and VQ/HMM 1 Behrouz Vasegh, 2 Somayeh Hashem, 1,2 Deparmen
More informationA Novel Discrete Differential Evolution Algoritnm for Task Scheduling in Heterogeneous Computing Systems
Proceedngs of he 2009 IEEE Inernaonal Conference on Sysems, Man, and Cybernecs San Anono, TX, USA - Ocober 2009 A Novel Dscree Dfferenal Evoluon Algornm for Task Schedulng n Heerogeneous Compung Sysems
More informationComparison of Differences between Power Means 1
In. Journal of Mah. Analyss, Vol. 7, 203, no., 5-55 Comparson of Dfferences beween Power Means Chang-An Tan, Guanghua Sh and Fe Zuo College of Mahemacs and Informaon Scence Henan Normal Unversy, 453007,
More informationPerformance Analysis for a Network having Standby Redundant Unit with Waiting in Repair
TECHNI Inernaonal Journal of Compung Scence Communcaon Technologes VOL.5 NO. July 22 (ISSN 974-3375 erformance nalyss for a Nework havng Sby edundan Un wh ang n epar Jendra Sngh 2 abns orwal 2 Deparmen
More information2/20/2013. EE 101 Midterm 2 Review
//3 EE Mderm eew //3 Volage-mplfer Model The npu ressance s he equalen ressance see when lookng no he npu ermnals of he amplfer. o s he oupu ressance. I causes he oupu olage o decrease as he load ressance
More informationApplication of thermal error in machine tools based on Dynamic. Bayesian Network
Inernaonal Journal of Research n Engneerng and Scence (IJRES) ISS (Onlne): 2320-9364, ISS (Prn): 2320-9356 www.res.org Volume 3 Issue 6 ǁ June 2015 ǁ PP.22-27 pplcaon of hermal error n machne ools based
More information5th International Conference on Advanced Design and Manufacturing Engineering (ICADME 2015)
5h Inernaonal onference on Advanced Desgn and Manufacurng Engneerng (IADME 5 The Falure Rae Expermenal Sudy of Specal N Machne Tool hunshan He, a, *, La Pan,b and Bng Hu 3,c,,3 ollege of Mechancal and
More informationDiscrete 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 informationBayesian Inference of the GARCH model with Rational Errors
0 Inernaonal Conference on Economcs, Busness and Markeng Managemen IPEDR vol.9 (0) (0) IACSIT Press, Sngapore Bayesan Inference of he GARCH model wh Raonal Errors Tesuya Takash + and Tng Tng Chen Hroshma
More informationIn the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!") i+1,q - [(!
ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL The frs hng o es n wo-way ANOVA: Is here neracon? "No neracon" means: The man effecs model would f. Ths n urn means: In he neracon plo (wh A on he horzonal
More informationAutomatic Driving System Based on Embedded Microcontroller Using Human Skill Models
JOURNAL OF COMPUTERS VOL. 6 NO. 2 FEBRUARY 20 35 Auomac Drvng Sysem Based on Embedded Mcroconroller Usng Human Skll Models Shangchang Ma Chengdu Unversy of Informaon Technology (CMA. Key Laboraory of Amospherc
More informationJ i-1 i. J i i+1. Numerical integration of the diffusion equation (I) Finite difference method. Spatial Discretization. Internal nodes.
umercal negraon of he dffuson equaon (I) Fne dfference mehod. Spaal screaon. Inernal nodes. R L V For hermal conducon le s dscree he spaal doman no small fne spans, =,,: Balance of parcles for an nernal
More informationBoosted LMS-based Piecewise Linear Adaptive Filters
016 4h European Sgnal Processng Conference EUSIPCO) Boosed LMS-based Pecewse Lnear Adapve Flers Darush Kar and Iman Marvan Deparmen of Elecrcal and Elecroncs Engneerng Blken Unversy, Ankara, Turkey {kar,
More informationDecentralised Sliding Mode Load Frequency Control for an Interconnected Power System with Uncertainties and Nonlinearities
Inernaonal Research Journal of Engneerng and echnology IRJE e-iss: 2395-0056 Volume: 03 Issue: 12 Dec -2016 www.re.ne p-iss: 2395-0072 Decenralsed Sldng Mode Load Frequency Conrol for an Inerconneced Power
More informationdoi: info:doi/ /
do: nfo:do/0.063/.322393 nernaonal Conference on Power Conrol and Opmzaon, Bal, ndonesa, -3, June 2009 A COLOR FEATURES-BASED METHOD FOR OBJECT TRACKNG EMPLOYNG A PARTCLE FLTER ALGORTHM Bud Sugand, Hyoungseop
More informationInversion of Complex Valued Neural Networks Using Complex Back-propagation Algorithm
ITERATIOAL JOURAL OF ATHEATICS AD COPUTERS I SIULATIO Inverson of Complex Valued eural ewors Usng Complex Bac-propagaon Algorhm Ana S. Gangal, P.K. Kalra, and D.S.Chauhan Absrac Ths paper presens he nverson
More informationABSTRACT. KEYWORDS Hybrid, Genetic Algorithm, Shipping, Dispatching, Vehicle, Time Windows INTRODUCTION
A HYBRID GENETIC ALGORITH FOR A DYNAIC INBOUND ORDERING AND SHIPPING AND OUTBOUND DISPATCHING PROBLE WITH HETEROGENEOUS VEHICLE TYPES AND DELIVERY TIE WINDOWS by Byung Soo Km, Woon-Seek Lee, and Young-Seok
More informationImproved Coupled Tank Liquid Levels System Based on Swarm Adaptive Tuning of Hybrid Proportional-Integral Neural Network Controller
Amercan J. of Engneerng and Appled Scences (4): 669-675, 009 ISSN 94-700 009 Scence Publcaons Improved Coupled Tan Lqud Levels Sysem Based on Swarm Adapve Tunng of Hybrd Proporonal-Inegral Neural Newor
More informationCHAPTER 10: LINEAR DISCRIMINATION
CHAPER : LINEAR DISCRIMINAION Dscrmnan-based Classfcaon 3 In classfcaon h K classes (C,C,, C k ) We defned dscrmnan funcon g j (), j=,,,k hen gven an es eample, e chose (predced) s class label as C f g
More informationSolution in semi infinite diffusion couples (error function analysis)
Soluon n sem nfne dffuson couples (error funcon analyss) Le us consder now he sem nfne dffuson couple of wo blocks wh concenraon of and I means ha, n a A- bnary sysem, s bondng beween wo blocks made of
More informationFuzzy Set Theory in Modeling Uncertainty Data. via Interpolation Rational Bezier Surface Function
Appled Mahemacal Scences, Vol. 7, 013, no. 45, 9 38 HIKARI Ld, www.m-hkar.com Fuzzy Se Theory n Modelng Uncerany Daa va Inerpolaon Raonal Bezer Surface Funcon Rozam Zakara Deparmen of Mahemacs, Faculy
More information( ) lamp power. dx dt T. Introduction to Compact Dynamical Modeling. III.1 Reducing Linear Time Invariant Systems
SF & IH Inroducon o Compac Dynamcal Modelng III. Reducng Lnear me Invaran Sysems Luca Danel Massachuses Insue of echnology Movaons dx A x( + b u( y( c x( Suppose: we are jus neresed n ermnal.e. npu/oupu
More information