Optimizing Back-Propagation using PSO_Hill and PSO_A*

Size: px
Start display at page:

Download "Optimizing Back-Propagation using PSO_Hill and PSO_A*"

Transcription

1 Inernaional Journal of Scienific and Research Publicaions, Volume 3, Issue 4, April Opimizing Back-Propagaion using PSO_Hill and PSO_A* Priyanka Sharma *, Asha Mishra ** * M.Tech Scholar of compuer science & Engineering, BSAITM, Faridabad ** Deparmen of compuer science & Engineering, BSAITM, Faridabad Absrac- Back propagaion algorihm (BPA) have he complexiy, local minima problem so we are using Paricle Swarm opimizaion (PSO) algorihms o reduce and opimize BPA. In his paper, wo varians of Paricle Swarm Opimizaion (PSO) PSO_Hill and PSO_A* is used as opimizaion algorihm. PSO_Hill and PSO_A* algorihms are analyzed and evaluaed on he basis of heir advanages, applied o feed forward neural nework(fnn) for back propagaion algorihm(bpa) which is a gredien desen echnique. where BPA is used for non_linear problems. These non_linear problems are improved by a PSO_Hill and PSO_A* algorihms. Index Terms- BPA, PSO_hill, PSO_A*, ANN O I. INTRODUCTION pimizaion is an acive area of research as many algorihms ha are inroduced earlier are complex and no opimize so beer opimizaion algorihms are needed. The objecive of opimizaion algorihm is o find a soluion o saisfying a se of consrains such ha objecive funcion is maximized or minimized. Paricle swarm opimizaion (PSO) is an alernaive populaion-based evoluionary compuaion echnique. I has been shown o be capable of opimizing hard mahemaical problems in coninuous or binary space. The paricle swarm opimizaion (PSO) algorihm is based on he evoluionary compuaion echnique [11-13]. PSO opimizes an objecive funcion by conducing populaion-based search. The populaion consiss of poenial soluions, called paricles, similar o birds in a flock. The paricles are randomly iniialized and hen freely fly across he muli-dimensional search space. While flying, every paricle updaes is velociy and posiion based on is own bes experience and ha of he enire populaion. The updaing policy will cause he paricle swarm o move oward a region wih a higher objec value. Evenually, all he paricles will gaher around he poin wih he highes objec value. PSO processes he search scheme using populaions of paricles where each paricle is equivalen o a candidae soluion of a problem [9]. The paricle moves according o an adjused velociy, which is based on ha paricle s experience and he experience of is companions. For he D-dimensional funcion f(.), he i h paricle for he j h ieraion can be represened as X j = X 1, X 2 X n W j= W 1, W 2 W n Assume ha he bes local posiion of he j h paricle a hen nh ieraion is represened as L_bes j =L_bes 1, L_bes 2. L_bes n The bes posiion amongs all he paricles, G_bes, from he firs ieraion o he j h ieraion, where bes is defined by some funcion of he swarm, is G_bes j =G_bes 1, G_bes2.G_bes n The original paricle swarm opimizaion algorihm can be expressed as follows: P j = X j *W j V j = f(p j ) means V j =1/(1+exp(-P j )) V j +1 = V j +r1* c1(l_bes j -X j )+r2*c2(g_bes j -X j ) X j +1 = X j + V i +1 r1and r 2 are random variables acs as learning signals such ha 0 r 1, r 2 1. where W j is he ineria weigh a he j h ieraion. The weighing facors, C 1 and C 2 are used as consans. In his paper, he opimizaion and analysis of back propagaion algorihm is done by paricle swarm opimizaion, and wo varian of paricle swarm opimizaion PSO_Hill and PSO_A*and heir algorihm, archiecure are proposed. This paper is organized as follows. In Secion II, we explained original swarm echniques ha are PSO and ACO. In Secion III, we explained original Back-propagaion nework wih algorihms. In secion IV, we explained Relaed work in PSO. In Secion V, we explained proposed work (PSO_Hill and PSO_A* varians wih diagram and algorihm). In secion VI, we explained Conclusion of paper. II. SWARM TECHNIQUES Swarm inelligence (SI) is he collecive behavior of decenralized, self-organized sysems, naural or arificial. The concep is employed in work on arificial inelligence. SI sysems are ypically made up of a populaion of simple agens ineracing locally wih one anoher and wih heir environmen. The inspiraion ofen comes from naure, especially biological sysems. The agens follow very simple rules, and alhough here

2 Inernaional Journal of Scienific and Research Publicaions, Volume 3, Issue 4, April is no cenralized conrol srucure dicaing how individual agens should behave, local, and o a cerain degree random, ineracions beween such agens lead o he emergence of "inelligen" global behavior, unknown o he individual agens. Naural examples of SI include an colonies, bird flocking, animal herding, bacerial growh, and fish schooling. Two well known approaches of swarm inelligence are: An Colony Opimizaion (ACO): An colony opimizaion (ACO) is a class of opimizaion algorihms modeled on he acions of an an colony [2], [13]. ACO mehods are useful in problems ha need o find pahs o goals. Arificial 'ans' simulaion agens locae opimal soluions by moving hrough a parameer space represening all possible soluions. Naural ans lay down pheromones direcing each oher o resources while exploring heir environmen. The simulaed 'ans' similarly record heir posiions and he qualiy of heir soluions, so ha in laer simulaion ieraions more ans locae beer soluions. Paricle Swarm Opimizaion (PSO): Paricle Swarm Opimizaion (PSO) is a swarm-based inelligence algorihm [6] influenced by he social behavior of animals such as a flock of birds finding a food source or a school of fish proecing hemselves from a predaor. A paricle in PSO is analogous o a bird or fish flying hrough a search (problem) space. The movemen of each paricle is co-coordinaed by a velociy which has boh magniude and direcion. Each paricle posiion a any insance of ime is influenced by is bes posiion and he posiion of he bes paricle in a problem space. The performance of a paricle is measured by a finess value, which is problem specific. Each paricle will have a finess value, which will be evaluaed by a finess funcion o be opimized in each generaion. Each paricle knows is bes posiion L_bes and he bes posiion so far among he enire group of paricles G_bes. The L_bes of a paricle is he bes resul (finess value) so far reached The paricle swarm opimizaion (PSO) algorihm is based on he evoluionary compuaion echnique. PSO opimizes an objecive funcion by conducing populaion-based search. The populaion consiss of poenial soluions, called paricles, similar o birds in a flock. The paricles are randomly iniialized and hen freely fly across he muli-dimensional search space. While flying, every paricle updaes is velociy and posiion based on is own bes experience and ha of he enire populaion. The updaing policy will cause he paricle swarm o move oward a region wih a higher objec value. Evenually, all he paricles will gaher around he poin wih he highes objec value. PSO aemps o simulae social behavior, which differs from he naural selecion schemes of geneic algorihms. PSO processes he search scheme using populaions of paricles, which corresponds o he use of individuals in geneic algorihms. Each paricle is equivalen o a candidae soluion of a problem. The paricle moves according o an adjused velociy, which is based on ha paricle s experience and he experience of is companions. A) Convenional Paricle Swarm Opimizaion The paricle swarm opimizaion algorihm was inroduced by Kennedy and Eberhar in 1995 [6], [13]. The algorihm consiss of a swarm of paricles flying hrough he search space. Each individual i in he swarm conains parameers for posiion x i and velociy v i, where x i R n, v i R n while n is he dimension of he search space. The posiion of each paricle represens a poenial soluion o he opimizaion problem. The dynamics of he swarm are governed by a se of rules ha modify he velociy of each paricle according o he experience of he paricle and by adding a velociy vecor o he curren posiion, he posiion of each paricle is modified. As he paricles move around he space, differen finess values are given o he paricles a differen locaions according o how he curren posiions of paricles saisfy he objecive. A each ieraion, each paricle keeps rack of is local bes posiion, L_bes and depending on he social nework srucure of he swarm, he global bes posiion, G_bes. B) Applicaion of PSO 1. Neural Nework Training 2. Telecommunicaions 3. Daa Mining 4. Design and Combinaorial opimizaion 5. Power sysems 6. Signal processing III. BACK-PROPAGATION ALGORITHM The back- propagaion algorihm is used in layered feedforward ANNs [14]. This means ha he arificial neurons are organized in layers, and send heir signals forward, and hen he errors are propagaed backwards. The nework receives inpus by neurons in he inpu layer, and he oupu of he nework is given by he neurons on an oupu layer. There may be one or more inermediae hidden layers. The backpropagaion algorihm uses supervised learning, which means ha we provide he algorihm wih examples of he inpus and oupus we wan he nework o compue, and hen he error (difference beween acual and expeced resuls) is calculaed. The idea of he back propagaion algorihm is o reduce his error, unil he ANN learns he raining daa. The raining begins wih random weighs, and he goal is o adjus hem so ha he error will be minimal. The basic back propagaion algorihm consiss of hree seps. 1) The inpu paern is presened o he inpu layer of he nework. These inpus are propagaed hrough he nework unil hey reach he oupu unis. This forward pass produces he acual or prediced oupu paern. 2) The acual nework oupus are subraced from he desired oupus and an error signal is produced. 3) This error signal is hen he basis for he back propagaion sep, whereby he errors are passed back hrough he neural nework by compuing he conribuion of each hidden processing uni and deriving he corresponding adjusmen needed o produce he correc oupu. The connecion weighs are hen adjused and he neural nework has jus learned from an experience. 4) Learning parameers are used o conrol he raining process of a back propagaion nework. 5) The learn rae is used o specify wheher he neural nework is going o make major adjusmens afer each

3 Inernaional Journal of Scienific and Research Publicaions, Volume 3, Issue 4, April learning rial or if i is only going o make minor adjusmens. 6) Momenum is used o conrol possible oscillaions in he weighs, which could be caused by alernaely signed error signals. IV. RELATED WORK Pillai, K. G. [1] explains a novel overlapping swarm inelligence algorihm is inroduced o rain he weighs of an arificial neural nework. Training a neural nework is a difficul ask ha requires an effecive search mehodology o compue he weighs along he edges of a nework. The back propagaion algorihm, a gradien based mehod, is frequenly used o rain mulilayer feed-forward neworks. On he oher hand, raining algorihms based on evoluionary compuaion have been used o rain mulilayer feed-forward neworks in an aemp o overcome he limiaions of gradien based algorihms wih mixed resuls. This paper inroduces an overlapping swarm inelligence echnique o rain mulilayer feedforward neworks. The resuls show ha OSI mehod performs eiher on par wih or beer han he oher mehods esed. M. Conforh and Y. Meng [2] propose a swarm inelligence based reinforcemen learning (SWIRL) mehod o rain arificial neural neworks (ANN). Basically, wo swarm inelligence based algorihms are combined ogeher o rain he ANN models. An Colony Opimizaion (ACO) is applied o selec ANN opology, while Paricle Swarm Opimizaion (PSO) is applied o adjus ANN connecion weighs. To evaluae he performance of he SWIRL model, i is applied o double pole problem and robo localizaion hrough reinforcemen learning. Exensive simulaion resuls successfully demonsrae ha SWIRL offers performance ha is compeiive wih modern neuro-evoluionary echniques, as well as is viabiliy for realworld problems. Marco Dorigo and Mauro Biraari [3] defines Swarm inelligence as he discipline ha deals wih naural and arificial sysems composed of many individuals ha coordinae using decenralized conrol and self-organizaion. In paricular, he discipline focuses on he collecive behaviors ha resul from he local ineracions of he individuals wih each oher and wih heir environmen. Examples of sysems sudied by swarm inelligence are colonies of ans and ermies, schools of fish, flocks of birds, herds of land animals. Some human arifacs also fall ino he domain of swarm inelligence, noably some mulirobo sysems, and also cerain compuer programs ha are wrien o ackle opimizaion and daa analysis problems. Y. Karpa and Tugrul Ozel [4] propose a concep of paricle swarm opimizaion, which is a recenly developed evoluionary algorihm, is used o opimize machining parameers in hard urning processes where muliple conflicing objecives are presen.the relaionships beween machining parameers and he performance measures of ineres are obained by using experimenal daa and swarm inelligen neural nework sysems (SINNS). The resuls showed ha paricle swarm opimizaion is an effecive mehod for solving muli-objecive opimizaion problems, and an inegraed sysem of neural neworks and swarm inelligence can be used in solving complex machining opimizaion problems. James kennedy and Russell Eberhar [5] propose a concep for he opimizaion of nonlinear funcions using paricle swarm mehodology is inroduced. The evoluion of several paradigms is oulined, and an implemenaion of one of he paradigms is discussed. Benchmark esing of he paradigm is described, and applicaions, including nonlinear funcion opimizaion and neural nework raining, are proposed. The relaionships beween paricle swarm opimizaion and boh arificial life and geneic algorihms are described. V. PROPOSED WORK 5.1) Analysis of Differen Algorihm: This is done on he basis of analysis of heir advanage o inroduce proposed varians for opimizaion. In his we analysis he limiaion of BPA and o opimize BPA we use PSO approach ) Disadvanages in BPA: a) Local Minima b) Low Speed c) Higher cos of compuaion d) Error Problem e) Gloal Maxima bu less as compared o Local f) Less accuracy 5.1.2) PSO Advanages: a) PSO can be applied o scienific research and Engineering. b) High compuaional speed c) Learning achieved from paricle own experienced d) Learning achieved from experience of cooperaion beween paricles 5.2) Proposed Varians for opimizaion: 1) PSO_hill 2) PSO_A* 5.2.1) PSO_hill Advange: a) High compuaional speed b) Srong abiliy in global search c) Higher Accuracy d) Learning achieved from paricle own experienced e) Learning achieved from experience of cooperaion beween paricles ) PSO_Hill variables Noaions: X: Iniial paricle posiion. W: Weigh for each paricle. n: Toal no. of ieraions. P curr = curren posiion of paricles P bes = paricle bes posiion

4 Inernaional Journal of Scienific and Research Publicaions, Volume 3, Issue 4, April ) PSO_Hill Diagram: Iniialize posiion X i, velociy and weigh w. Iniialize posiion X j, velociy and weigh w. P j=x j*w j P j= W j* X j Vj= 1/(1+exp(-v j)) V j = 1/(1+exp(-p j)) P j=goal j<n Goal=V j Selec bes among paricle, compue p_global YE S Curren is beer han Searched Sae P_global=p_goal Se p curr=p bes N O Compue finess funcion Opimized Opimized Sop Sop 5.2.4) PSO_Hill Algorihm: 1) Sar wih iniializing paricle posiion X, heir velociy V, Weigh W, p_loc 0, p_glob 0,n 0 2) If goal=v j hen erminae i; Oherwise 3) P j =X j *W j 4) Vj=1/(1+exp(-p j )) 5) If (j<n) 6) { 7) If(Evaluae a new posiion which is beer han curren posiion bu no goal) 8) { 9) Curr_posiion=beer_posiion; 10) Else 11) Keep curren posiion & coninue he search o find goal 12) } 13) Else 14) Reurn o sep 2 o coninue ill i reaches o goal 5.3.1) PSO_A* Advanages: I. Mos heurisic soluion II. Opimized in erms of finess funcion 5.3.2) PSO_A* Diagram: 5.2.3) PSO_A* Variables Noaions: Open_lis: A lis which have nodes ha are generaed bu no expanded. Closed_lis: A lis which have nodes ha are expanded and is childrens are available o search program ) PSO_A* Algorihm: 1. Sar wih iniial posiion of paricles and place hem on open node, p_loc 0,p_glob If (open_lis=empy) sop and reurn as failure 3. Selec p_loc paricles n from open lis ha has smalles finess funcion 4. { 5. if node n= goal node 6. reurn success 7. sop 8. } 9. Oherwise 10. Expand he successor paricles of node n and as node n is explored so keep i on closed lis 11. For each successor h 12. { 13. If h is no in open_lis or closed_lis 14. { 15. Aach a back poiner o n paricle o backrack i 16. {

5 Inernaional Journal of Scienific and Research Publicaions, Volume 3, Issue 4, April Compue finess of paricle f*(h) 18. Else 19. Compue lowes or p_glob (g*(h)) 20. } 21. Place on open lis 22. Reurn o Sep o ill he goal VI. CONCLUSION In his paper, wo varians of he paricle swarm opimizaion scheme is presened. Two PSO Varians PSO_Hill and PSO_A* are proposed wih heir algorihm, archiecure, advanages and disadvanages, which can be used o he opimize he BPA. In nex paper, a hird sraegy is proposed PSO_Hill_A* on he basis of srengh of wo varians PSO_Hill and PSO_A* algorihm. The paricle local bes and global bes posiions help he varians o move owards he soluion. REFERENCES [1] Pillai, K. G, Overlapping swarm inelligence wih arificial neural nework,swarm inelligence(sis),2011 IEEE Symopsium,15 april [2] Mahew Conforh and Yan Meng, Reinforcemen Learning for Neural Neworks using Swarm Inelligence, 2008 IEEE Swarm Inelligence Symposium, S. Louis MO USA, Sepember 21-23, 2008 [3] [3] Macro Dorigo and Mauro Biraari (2007), Scholarpedia, 2(9) Swarm inelligence, hp:// Marco Dorigo (2007) An colony opimizaion. Scholarpedia, 2(3):1461. [4] Yiği Karpa and Tuğrul Özel, Swarm-Inelligen Neural Nework Sysem (SINNS) Based Muli-Objecive Opimizaion Of Hard Turning,Transacions of NAMRI/SME Volume 34, [5] James Kennedy' and Russell Eberhar2, Paricle Swarm Opimizaion,, Washingon, DC 20212,kennedy_jim@bls.gov,hp:// orks/hw3/_reading6%201995%20paricle%20swarming.pdf, 1995 IEEE. [6].Kennedy and R.Eberhar, Paricle swarm opimizaion,proceedings of. IEEE Inernaional Conference on Neural Neworks, pp , 1995 [7] Boonserm Kaewkamnerdpong and Peer J. Benley, percepive Paricle Swarm Opimizaion:An Invesigaion [8] Jui-Fang Chang, Shu-Chuan Chu, John F. Roddick and Jeng-Shyang Pan, A Parallel Paricle Swarm Opimizaion Algorihm wih Communicaion Sraegies, Journal of Informaion Science and Engineering 21, (2005) [9] R. Eberhar and J. Kennedy, A new opimizer using paricle swarm heory in Proceedings of 6h Inernaional Symposium on Micro Machine and Human Science, 1995, pp [10] J. Kennedy and R. Eberhar, Paricle swarm opimizaion, in Proceedings of IEEE Inernaional Conference on Neural Neworks, 1995, pp [11] P. Tarasewich and P. R. McMullen, Swarm inelligence, Communicaions of he ACM, Vol. 45, 2002, pp [12] Niu Mahuriya and Dr. Ashish Bansal, Applicabiliy of back propagaion neural nework for recruimen daa mining, Inernaional Journal of Engineering Research & Technology (IJERT), ISSN: , Vol. 1 Issue 3, May [13] V.Selvi and Dr.R.Umarani, Comparaive Analysis of An Colony and Paricle Swarm Opimizaion Techniques, Inernaional Journal of Compuer Applicaions ( ), Volume 5 No.4, Augus 2010 [14] Carlos Gershenson, Arificial Neural Neworks for Beginners AUTHORS Firs Auhor Priyanka Sharma received B.TECH degree in Compuer Science and Engineering wih Hons. from Maharshi Dayanand Universiy in 2011 and is persuing M.Tech. in Compuer Engineering. Presenly, She is working as Lecurer in Compuer Engineering deparmen in RITM, Palwal. Her areas of ineress are Arificial Neural Nework, Sof Compuing, Paern Recogniion., priya.sharma090@gmail.com Second Auhor Asha Mishra received B.E. degree in Compuer Science & Engineering from NIT, Assam and M.Tech in Compuer Science from A.I.T.M, Palwal. Presenly, she is working as Senior Lecurer in Compuer Engineering deparmen in B.S.A. Insiue of Technology & Managemen, Faridabad. Her areas of ineress are DBMS, Analysis & Design of Algorihm and Nework Securiy., asha1.mishra@gmail.com

Particle Swarm Optimization

Particle Swarm Optimization Paricle Swarm Opimizaion Speaker: Jeng-Shyang Pan Deparmen of Elecronic Engineering, Kaohsiung Universiy of Applied Science, Taiwan Email: jspan@cc.kuas.edu.w 7/26/2004 ppso 1 Wha is he Paricle Swarm Opimizaion

More information

CHAPTER 10 VALIDATION OF TEST WITH ARTIFICAL NEURAL NETWORK

CHAPTER 10 VALIDATION OF TEST WITH ARTIFICAL NEURAL NETWORK 175 CHAPTER 10 VALIDATION OF TEST WITH ARTIFICAL NEURAL NETWORK 10.1 INTRODUCTION Amongs he research work performed, he bes resuls of experimenal work are validaed wih Arificial Neural Nework. From he

More information

Air Quality Index Prediction Using Error Back Propagation Algorithm and Improved Particle Swarm Optimization

Air Quality Index Prediction Using Error Back Propagation Algorithm and Improved Particle Swarm Optimization Air Qualiy Index Predicion Using Error Back Propagaion Algorihm and Improved Paricle Swarm Opimizaion Jia Xu ( ) and Lang Pei College of Compuer Science, Wuhan Qinchuan Universiy, Wuhan, China 461406563@qq.com

More information

Particle Swarm Optimization Combining Diversification and Intensification for Nonlinear Integer Programming Problems

Particle Swarm Optimization Combining Diversification and Intensification for Nonlinear Integer Programming Problems Paricle Swarm Opimizaion Combining Diversificaion and Inensificaion for Nonlinear Ineger Programming Problems Takeshi Masui, Masaoshi Sakawa, Kosuke Kao and Koichi Masumoo Hiroshima Universiy 1-4-1, Kagamiyama,

More information

STATE-SPACE MODELLING. A mass balance across the tank gives:

STATE-SPACE MODELLING. A mass balance across the tank gives: B. Lennox and N.F. Thornhill, 9, Sae Space Modelling, IChemE Process Managemen and Conrol Subjec Group Newsleer STE-SPACE MODELLING Inroducion: Over he pas decade or so here has been an ever increasing

More information

An introduction to the theory of SDDP algorithm

An introduction to the theory of SDDP algorithm An inroducion o he heory of SDDP algorihm V. Leclère (ENPC) Augus 1, 2014 V. Leclère Inroducion o SDDP Augus 1, 2014 1 / 21 Inroducion Large scale sochasic problem are hard o solve. Two ways of aacking

More information

Ensamble methods: Bagging and Boosting

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

More information

STUDY ON VELOCITY CLAMPING IN PSO USING CEC`13 BENCHMARK

STUDY ON VELOCITY CLAMPING IN PSO USING CEC`13 BENCHMARK STUDY ON VELOCITY CLAMPING IN PSO USING CEC`13 BENCHMARK Michal Pluhacek, Roman Senkerik, Adam Vikorin and Tomas Kadavy Tomas Baa Universiy in Zlin, Faculy of Applied Informaics Nam T.G. Masaryka 5555,

More information

CSE/NB 528 Lecture 14: Reinforcement Learning (Chapter 9)

CSE/NB 528 Lecture 14: Reinforcement Learning (Chapter 9) CSE/NB 528 Lecure 14: Reinforcemen Learning Chaper 9 Image from hp://clasdean.la.asu.edu/news/images/ubep2001/neuron3.jpg Lecure figures are from Dayan & Abbo s book hp://people.brandeis.edu/~abbo/book/index.hml

More information

Ensamble methods: Boosting

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

More information

Applying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Capacity Constraints

Applying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Capacity Constraints IJCSI Inernaional Journal of Compuer Science Issues, Vol 9, Issue 1, No 1, January 2012 wwwijcsiorg 18 Applying Geneic Algorihms for Invenory Lo-Sizing Problem wih Supplier Selecion under Sorage Capaciy

More information

A Hop Constrained Min-Sum Arborescence with Outage Costs

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

More information

MODULE - 9 LECTURE NOTES 2 GENETIC ALGORITHMS

MODULE - 9 LECTURE NOTES 2 GENETIC ALGORITHMS 1 MODULE - 9 LECTURE NOTES 2 GENETIC ALGORITHMS INTRODUCTION Mos real world opimizaion problems involve complexiies like discree, coninuous or mixed variables, muliple conflicing objecives, non-lineariy,

More information

A Framework for Efficient Document Ranking Using Order and Non Order Based Fitness Function

A Framework for Efficient Document Ranking Using Order and Non Order Based Fitness Function A Framework for Efficien Documen Ranking Using Order and Non Order Based Finess Funcion Hazra Imran, Adii Sharan Absrac One cenral problem of informaion rerieval is o deermine he relevance of documens

More information

Modal identification of structures from roving input data by means of maximum likelihood estimation of the state space model

Modal identification of structures from roving input data by means of maximum likelihood estimation of the state space model Modal idenificaion of srucures from roving inpu daa by means of maximum likelihood esimaion of he sae space model J. Cara, J. Juan, E. Alarcón Absrac The usual way o perform a forced vibraion es is o fix

More information

Physics 235 Chapter 2. Chapter 2 Newtonian Mechanics Single Particle

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

More information

Optimal Design of LQR Weighting Matrices based on Intelligent Optimization Methods

Optimal Design of LQR Weighting Matrices based on Intelligent Optimization Methods Opimal Design of LQR Weighing Marices based on Inelligen Opimizaion Mehods Inernaional Journal of Inelligen Informaion Processing, Volume, Number, March Opimal Design of LQR Weighing Marices based on Inelligen

More information

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

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

More information

Single-Pass-Based Heuristic Algorithms for Group Flexible Flow-shop Scheduling Problems

Single-Pass-Based Heuristic Algorithms for Group Flexible Flow-shop Scheduling Problems Single-Pass-Based Heurisic Algorihms for Group Flexible Flow-shop Scheduling Problems PEI-YING HUANG, TZUNG-PEI HONG 2 and CHENG-YAN KAO, 3 Deparmen of Compuer Science and Informaion Engineering Naional

More information

Robust and Learning Control for Complex Systems

Robust and Learning Control for Complex Systems Robus and Learning Conrol for Complex Sysems Peer M. Young Sepember 13, 2007 & Talk Ouline Inroducion Robus Conroller Analysis and Design Theory Experimenal Applicaions Overview MIMO Robus HVAC Conrol

More information

Resource Allocation in Visible Light Communication Networks NOMA vs. OFDMA Transmission Techniques

Resource Allocation in Visible Light Communication Networks NOMA vs. OFDMA Transmission Techniques Resource Allocaion in Visible Ligh Communicaion Neworks NOMA vs. OFDMA Transmission Techniques Eirini Eleni Tsiropoulou, Iakovos Gialagkolidis, Panagiois Vamvakas, and Symeon Papavassiliou Insiue of Communicaions

More information

Vehicle Arrival Models : Headway

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

More information

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

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

More information

Some Basic Information about M-S-D Systems

Some Basic Information about M-S-D Systems Some Basic Informaion abou M-S-D Sysems 1 Inroducion We wan o give some summary of he facs concerning unforced (homogeneous) and forced (non-homogeneous) models for linear oscillaors governed by second-order,

More information

From Particles to Rigid Bodies

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

More information

Applying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Space

Applying Genetic Algorithms for Inventory Lot-Sizing Problem with Supplier Selection under Storage Space Inernaional Journal of Indusrial and Manufacuring Engineering Applying Geneic Algorihms for Invenory Lo-Sizing Problem wih Supplier Selecion under Sorage Space Vichai Rungreunganaun and Chirawa Woarawichai

More information

Application of a Stochastic-Fuzzy Approach to Modeling Optimal Discrete Time Dynamical Systems by Using Large Scale Data Processing

Application of a Stochastic-Fuzzy Approach to Modeling Optimal Discrete Time Dynamical Systems by Using Large Scale Data Processing Applicaion of a Sochasic-Fuzzy Approach o Modeling Opimal Discree Time Dynamical Sysems by Using Large Scale Daa Processing AA WALASZE-BABISZEWSA Deparmen of Compuer Engineering Opole Universiy of Technology

More information

Dimitri Solomatine. D.P. Solomatine. Data-driven modelling (part 2). 2

Dimitri Solomatine. D.P. Solomatine. Data-driven modelling (part 2). 2 Daa-driven modelling. Par. Daa-driven Arificial di Neural modelling. Newors Par Dimiri Solomaine Arificial neural newors D.P. Solomaine. Daa-driven modelling par. 1 Arificial neural newors ANN: main pes

More information

Robust Learning Control with Application to HVAC Systems

Robust Learning Control with Application to HVAC Systems Robus Learning Conrol wih Applicaion o HVAC Sysems Naional Science Foundaion & Projec Invesigaors: Dr. Charles Anderson, CS Dr. Douglas Hile, ME Dr. Peer Young, ECE Mechanical Engineering Compuer Science

More information

Particle Swarm Optimization Algorithm for Agent-Based Artificial Markets. Tong Zhang and B. Wade Brorsen

Particle Swarm Optimization Algorithm for Agent-Based Artificial Markets. Tong Zhang and B. Wade Brorsen Paricle Swarm Opimizaion Algorihm for Agen-Based Arificial Markes Tong Zhang and B. Wade Brorsen Absrac Paricle swarm opimizaion (PSO) is adaped o simulae dynamic economic games. The robusness and speed

More information

EE 315 Notes. Gürdal Arslan CLASS 1. (Sections ) What is a signal?

EE 315 Notes. Gürdal Arslan CLASS 1. (Sections ) What is a signal? EE 35 Noes Gürdal Arslan CLASS (Secions.-.2) Wha is a signal? In his class, a signal is some funcion of ime and i represens how some physical quaniy changes over some window of ime. Examples: velociy of

More information

Presentation Overview

Presentation Overview Acion Refinemen in Reinforcemen Learning by Probabiliy Smoohing By Thomas G. Dieerich & Didac Busques Speaer: Kai Xu Presenaion Overview Bacground The Probabiliy Smoohing Mehod Experimenal Sudy of Acion

More information

KINEMATICS IN ONE DIMENSION

KINEMATICS IN ONE DIMENSION KINEMATICS IN ONE DIMENSION PREVIEW Kinemaics is he sudy of how hings move how far (disance and displacemen), how fas (speed and velociy), and how fas ha how fas changes (acceleraion). We say ha an objec

More information

Numerical Dispersion

Numerical Dispersion eview of Linear Numerical Sabiliy Numerical Dispersion n he previous lecure, we considered he linear numerical sabiliy of boh advecion and diffusion erms when approimaed wih several spaial and emporal

More information

Electrical and current self-induction

Electrical and current self-induction Elecrical and curren self-inducion F. F. Mende hp://fmnauka.narod.ru/works.hml mende_fedor@mail.ru Absrac The aricle considers he self-inducance of reacive elemens. Elecrical self-inducion To he laws of

More information

1 Review of Zero-Sum Games

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

More information

A Modified Particle Swarm Optimization For Engineering Constrained Optimization Problems

A Modified Particle Swarm Optimization For Engineering Constrained Optimization Problems Inernaional Journal of Compuer Science and Elecronics Engineering (IJCSEE Volume 3, Issue 1 (015 ISSN 30 408 (Online A Modified Paricle Swarm Opimizaion For Engineering Consrained Opimizaion Problems Choosak

More information

Ensemble Confidence Estimates Posterior Probability

Ensemble Confidence Estimates Posterior Probability Ensemble Esimaes Poserior Probabiliy Michael Muhlbaier, Aposolos Topalis, and Robi Polikar Rowan Universiy, Elecrical and Compuer Engineering, Mullica Hill Rd., Glassboro, NJ 88, USA {muhlba6, opali5}@sudens.rowan.edu

More information

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

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

More information

Multi-area Load Frequency Control using IP Controller Tuned by Particle Swarm Optimization

Multi-area Load Frequency Control using IP Controller Tuned by Particle Swarm Optimization esearch Journal of Applied Sciences, Engineering and echnology (): 96-, ISSN: -767 axwell Scienific Organizaion, Submied: July, Acceped: Sepember 8, Published: ecember 6, uli-area Load Frequency Conrol

More information

The Rosenblatt s LMS algorithm for Perceptron (1958) is built around a linear neuron (a neuron with a linear

The Rosenblatt s LMS algorithm for Perceptron (1958) is built around a linear neuron (a neuron with a linear In The name of God Lecure4: Percepron and AALIE r. Majid MjidGhoshunih Inroducion The Rosenbla s LMS algorihm for Percepron 958 is buil around a linear neuron a neuron ih a linear acivaion funcion. Hoever,

More information

Linear Response Theory: The connection between QFT and experiments

Linear Response Theory: The connection between QFT and experiments Phys540.nb 39 3 Linear Response Theory: The connecion beween QFT and experimens 3.1. Basic conceps and ideas Q: How do we measure he conduciviy of a meal? A: we firs inroduce a weak elecric field E, and

More information

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

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

More information

Keywords Digital Infinite-Impulse Response (IIR) filter, Digital Finite-Impulse Response (FIR) filter, DE, exploratory move

Keywords Digital Infinite-Impulse Response (IIR) filter, Digital Finite-Impulse Response (FIR) filter, DE, exploratory move Volume 5, Issue 7, July 2015 ISSN: 2277 128X Inernaional Journal of Advanced Research in Compuer Science and Sofware Engineering Research Paper Available online a: www.ijarcsse.com A Hybrid Differenial

More information

CSE/NB 528 Lecture 14: From Supervised to Reinforcement Learning (Chapter 9) R. Rao, 528: Lecture 14

CSE/NB 528 Lecture 14: From Supervised to Reinforcement Learning (Chapter 9) R. Rao, 528: Lecture 14 CSE/NB 58 Lecure 14: From Supervised o Reinforcemen Learning Chaper 9 1 Recall from las ime: Sigmoid Neworks Oupu v T g w u g wiui w Inpu nodes u = u 1 u u 3 T i Sigmoid oupu funcion: 1 g a 1 a e 1 ga

More information

Deep Learning: Theory, Techniques & Applications - Recurrent Neural Networks -

Deep Learning: Theory, Techniques & Applications - Recurrent Neural Networks - Deep Learning: Theory, Techniques & Applicaions - Recurren Neural Neworks - Prof. Maeo Maeucci maeo.maeucci@polimi.i Deparmen of Elecronics, Informaion and Bioengineering Arificial Inelligence and Roboics

More information

Machine Loading in Flexible Manufacturing System: A Swarm Optimization Approach

Machine Loading in Flexible Manufacturing System: A Swarm Optimization Approach Machine Loading in Flexible Manufacuring Sysem: A Swarm Opimizaion Approach Sandhyarani Biswas Naional Insiue of Technology sandhya_biswas@yahoo.co.in S.S.Mahapara Naional Insiue of Technology mahaparass2003@yahoo.com

More information

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

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

More information

An Optimal Dynamic Generation Scheduling for a Wind-Thermal Power System *

An Optimal Dynamic Generation Scheduling for a Wind-Thermal Power System * Energy and Power Engineering, 2013, 5, 1016-1021 doi:10.4236/epe.2013.54b194 Published Online July 2013 (hp://www.scirp.org/journal/epe) An Opimal Dynamic Generaion Scheduling for a Wind-Thermal Power

More information

On Comparison between Evolutionary Programming Network based Learning and Novel Evolution Strategy Algorithm-based Learning

On Comparison between Evolutionary Programming Network based Learning and Novel Evolution Strategy Algorithm-based Learning Proceedings of he ICEECE December -4 (003), Dhaka, Bangladesh On Comparison beween Evoluionary Programming Nework based Learning and vel Evoluion Sraegy Algorihm-based Learning M. A. Khayer Azad, Md. Shafiqul

More information

Sliding Mode Extremum Seeking Control for Linear Quadratic Dynamic Game

Sliding Mode Extremum Seeking Control for Linear Quadratic Dynamic Game Sliding Mode Exremum Seeking Conrol for Linear Quadraic Dynamic Game Yaodong Pan and Ümi Özgüner ITS Research Group, AIST Tsukuba Eas Namiki --, Tsukuba-shi,Ibaraki-ken 5-856, Japan e-mail: pan.yaodong@ais.go.jp

More information

Sequential Importance Resampling (SIR) Particle Filter

Sequential Importance Resampling (SIR) Particle Filter Paricle Filers++ Pieer Abbeel UC Berkeley EECS Many slides adaped from Thrun, Burgard and Fox, Probabilisic Roboics 1. Algorihm paricle_filer( S -1, u, z ): 2. Sequenial Imporance Resampling (SIR) Paricle

More information

Recent Developments In Evolutionary Data Assimilation And Model Uncertainty Estimation For Hydrologic Forecasting Hamid Moradkhani

Recent Developments In Evolutionary Data Assimilation And Model Uncertainty Estimation For Hydrologic Forecasting Hamid Moradkhani Feb 6-8, 208 Recen Developmens In Evoluionary Daa Assimilaion And Model Uncerainy Esimaion For Hydrologic Forecasing Hamid Moradkhani Cener for Complex Hydrosysems Research Deparmen of Civil, Consrucion

More information

Random Walk with Anti-Correlated Steps

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

More information

Lecture 9: September 25

Lecture 9: September 25 0-725: Opimizaion Fall 202 Lecure 9: Sepember 25 Lecurer: Geoff Gordon/Ryan Tibshirani Scribes: Xuezhi Wang, Subhodeep Moira, Abhimanu Kumar Noe: LaTeX emplae couresy of UC Berkeley EECS dep. Disclaimer:

More information

RC, RL and RLC circuits

RC, RL and RLC circuits Name Dae Time o Complee h m Parner Course/ Secion / Grade RC, RL and RLC circuis Inroducion In his experimen we will invesigae he behavior of circuis conaining combinaions of resisors, capaciors, and inducors.

More information

Robotics I. April 11, The kinematics of a 3R spatial robot is specified by the Denavit-Hartenberg parameters in Tab. 1.

Robotics I. April 11, The kinematics of a 3R spatial robot is specified by the Denavit-Hartenberg parameters in Tab. 1. Roboics I April 11, 017 Exercise 1 he kinemaics of a 3R spaial robo is specified by he Denavi-Harenberg parameers in ab 1 i α i d i a i θ i 1 π/ L 1 0 1 0 0 L 3 0 0 L 3 3 able 1: able of DH parameers of

More information

The field of mathematics has made tremendous impact on the study of

The field of mathematics has made tremendous impact on the study of A Populaion Firing Rae Model of Reverberaory Aciviy in Neuronal Neworks Zofia Koscielniak Carnegie Mellon Universiy Menor: Dr. G. Bard Ermenrou Universiy of Pisburgh Inroducion: The field of mahemaics

More information

Mechanical Fatigue and Load-Induced Aging of Loudspeaker Suspension. Wolfgang Klippel,

Mechanical Fatigue and Load-Induced Aging of Loudspeaker Suspension. Wolfgang Klippel, Mechanical Faigue and Load-Induced Aging of Loudspeaker Suspension Wolfgang Klippel, Insiue of Acousics and Speech Communicaion Dresden Universiy of Technology presened a he ALMA Symposium 2012, Las Vegas

More information

Section 3.5 Nonhomogeneous Equations; Method of Undetermined Coefficients

Section 3.5 Nonhomogeneous Equations; Method of Undetermined Coefficients Secion 3.5 Nonhomogeneous Equaions; Mehod of Undeermined Coefficiens Key Terms/Ideas: Linear Differenial operaor Nonlinear operaor Second order homogeneous DE Second order nonhomogeneous DE Soluion o homogeneous

More information

Stochastic Model for Cancer Cell Growth through Single Forward Mutation

Stochastic Model for Cancer Cell Growth through Single Forward Mutation Journal of Modern Applied Saisical Mehods Volume 16 Issue 1 Aricle 31 5-1-2017 Sochasic Model for Cancer Cell Growh hrough Single Forward Muaion Jayabharahiraj Jayabalan Pondicherry Universiy, jayabharahi8@gmail.com

More information

Forecasting Stock Exchange Movements Using Artificial Neural Network Models and Hybrid Models

Forecasting Stock Exchange Movements Using Artificial Neural Network Models and Hybrid Models Forecasing Sock Exchange Movemens Using Arificial Neural Nework Models and Hybrid Models Erkam GÜREEN and Gülgün KAYAKUTLU Isanbul Technical Universiy, Deparmen of Indusrial Engineering, Maçka, 34367 Isanbul,

More information

Georey E. Hinton. University oftoronto. Technical Report CRG-TR February 22, Abstract

Georey E. Hinton. University oftoronto.   Technical Report CRG-TR February 22, Abstract Parameer Esimaion for Linear Dynamical Sysems Zoubin Ghahramani Georey E. Hinon Deparmen of Compuer Science Universiy oftorono 6 King's College Road Torono, Canada M5S A4 Email: zoubin@cs.orono.edu Technical

More information

Chapter 21. Reinforcement Learning. The Reinforcement Learning Agent

Chapter 21. Reinforcement Learning. The Reinforcement Learning Agent CSE 47 Chaper Reinforcemen Learning The Reinforcemen Learning Agen Agen Sae u Reward r Acion a Enironmen CSE AI Faculy Why reinforcemen learning Programming an agen o drie a car or fly a helicoper is ery

More information

Open loop vs Closed Loop. Example: Open Loop. Example: Feedforward Control. Advanced Control I

Open loop vs Closed Loop. Example: Open Loop. Example: Feedforward Control. Advanced Control I Open loop vs Closed Loop Advanced I Moor Command Movemen Overview Open Loop vs Closed Loop Some examples Useful Open Loop lers Dynamical sysems CPG (biologically inspired ), Force Fields Feedback conrol

More information

Sliding Mode Controller for Unstable Systems

Sliding Mode Controller for Unstable Systems S. SIVARAMAKRISHNAN e al., Sliding Mode Conroller for Unsable Sysems, Chem. Biochem. Eng. Q. 22 (1) 41 47 (28) 41 Sliding Mode Conroller for Unsable Sysems S. Sivaramakrishnan, A. K. Tangirala, and M.

More information

θ with respect to time is

θ with respect to time is From MEC '05 Inergraing Prosheics and Medicine, Proceedings of he 005 MyoElecric Conrols/Powered Prosheics Symposium, held in Fredericon, New Brunswick, Canada, Augus 17-19, 005. A MINIMAL JERK PROSTHESIS

More information

Lab 10: RC, RL, and RLC Circuits

Lab 10: RC, RL, and RLC Circuits Lab 10: RC, RL, and RLC Circuis In his experimen, we will invesigae he behavior of circuis conaining combinaions of resisors, capaciors, and inducors. We will sudy he way volages and currens change in

More information

Embedded Systems and Software. A Simple Introduction to Embedded Control Systems (PID Control)

Embedded Systems and Software. A Simple Introduction to Embedded Control Systems (PID Control) Embedded Sysems and Sofware A Simple Inroducion o Embedded Conrol Sysems (PID Conrol) Embedded Sysems and Sofware, ECE:3360. The Universiy of Iowa, 2016 Slide 1 Acknowledgemens The maerial in his lecure

More information

Simulation-Solving Dynamic Models ABE 5646 Week 2, Spring 2010

Simulation-Solving Dynamic Models ABE 5646 Week 2, Spring 2010 Simulaion-Solving Dynamic Models ABE 5646 Week 2, Spring 2010 Week Descripion Reading Maerial 2 Compuer Simulaion of Dynamic Models Finie Difference, coninuous saes, discree ime Simple Mehods Euler Trapezoid

More information

Weightless Swarm Algorithm (WSA) for Dynamic Optimization Problems

Weightless Swarm Algorithm (WSA) for Dynamic Optimization Problems Weighless Swarm Algorihm (WSA) for Dynamic Opimizaion Problems T.O. Ting 1,*, Ka Lok Man 2, Sheng-Uei Guan 2, Mohamed Nayel 1, and Kaiyu Wan 2 1 Dep. Elecrical and Elecronic Eng. 2 Dep. Compuer Science

More information

Using the Kalman filter Extended Kalman filter

Using the Kalman filter Extended Kalman filter Using he Kalman filer Eended Kalman filer Doz. G. Bleser Prof. Sricker Compuer Vision: Objec and People Tracking SA- Ouline Recap: Kalman filer algorihm Using Kalman filers Eended Kalman filer algorihm

More information

Multi-scale 2D acoustic full waveform inversion with high frequency impulsive source

Multi-scale 2D acoustic full waveform inversion with high frequency impulsive source Muli-scale D acousic full waveform inversion wih high frequency impulsive source Vladimir N Zubov*, Universiy of Calgary, Calgary AB vzubov@ucalgaryca and Michael P Lamoureux, Universiy of Calgary, Calgary

More information

Rev. Téc. Ing. Univ. Zulia. Vol. 39, Nº 1, , 2016

Rev. Téc. Ing. Univ. Zulia. Vol. 39, Nº 1, , 2016 Rev. Téc. Ing. Univ. Zulia. Vol. 39, Nº 1, 358-363, 216 doi:1.21311/1.39.1.41 Face Deecion and Recogniion Based on an Improved Adaboos Algorihm and Neural Nework Haoian Zhang*, Jiajia Xing, Muian Zhu,

More information

In this chapter the model of free motion under gravity is extended to objects projected at an angle. When you have completed it, you should

In this chapter the model of free motion under gravity is extended to objects projected at an angle. When you have completed it, you should Cambridge Universiy Press 978--36-60033-7 Cambridge Inernaional AS and A Level Mahemaics: Mechanics Coursebook Excerp More Informaion Chaper The moion of projeciles In his chaper he model of free moion

More information

Chapter 2. First Order Scalar Equations

Chapter 2. First Order Scalar Equations Chaper. Firs Order Scalar Equaions We sar our sudy of differenial equaions in he same way he pioneers in his field did. We show paricular echniques o solve paricular ypes of firs order differenial equaions.

More information

Convergence Improvement of Reliability-Based MultiobjectiveOptimization Using Hybrid MOPSO

Convergence Improvement of Reliability-Based MultiobjectiveOptimization Using Hybrid MOPSO 10 h World Congress on Srucural and Mulidisciplinary Opimizaion May 19-4, 013, Orlando, Florida, USA Convergence Improvemen of Reliabiliy-Based MuliobjeciveOpimizaion Using Hybrid MOPSO Shoichiro KAWAJI

More information

m = 41 members n = 27 (nonfounders), f = 14 (founders) 8 markers from chromosome 19

m = 41 members n = 27 (nonfounders), f = 14 (founders) 8 markers from chromosome 19 Sequenial Imporance Sampling (SIS) AKA Paricle Filering, Sequenial Impuaion (Kong, Liu, Wong, 994) For many problems, sampling direcly from he arge disribuion is difficul or impossible. One reason possible

More information

) were both constant and we brought them from under the integral.

) were both constant and we brought them from under the integral. YIELD-PER-RECRUIT (coninued The yield-per-recrui model applies o a cohor, bu we saw in he Age Disribuions lecure ha he properies of a cohor do no apply in general o a collecion of cohors, which is wha

More information

1 Differential Equation Investigations using Customizable

1 Differential Equation Investigations using Customizable Differenial Equaion Invesigaions using Cusomizable Mahles Rober Decker The Universiy of Harford Absrac. The auhor has developed some plaform independen, freely available, ineracive programs (mahles) for

More information

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

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

More information

Orientation. Connections between network coding and stochastic network theory. Outline. Bruce Hajek. Multicast with lost packets

Orientation. Connections between network coding and stochastic network theory. Outline. Bruce Hajek. Multicast with lost packets Connecions beween nework coding and sochasic nework heory Bruce Hajek Orienaion On Thursday, Ralf Koeer discussed nework coding: coding wihin he nework Absrac: Randomly generaed coded informaion blocks

More information

On Measuring Pro-Poor Growth. 1. On Various Ways of Measuring Pro-Poor Growth: A Short Review of the Literature

On Measuring Pro-Poor Growth. 1. On Various Ways of Measuring Pro-Poor Growth: A Short Review of the Literature On Measuring Pro-Poor Growh 1. On Various Ways of Measuring Pro-Poor Growh: A Shor eview of he Lieraure During he pas en years or so here have been various suggesions concerning he way one should check

More information

The expectation value of the field operator.

The expectation value of the field operator. The expecaion value of he field operaor. Dan Solomon Universiy of Illinois Chicago, IL dsolom@uic.edu June, 04 Absrac. Much of he mahemaical developmen of quanum field heory has been in suppor of deermining

More information

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

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

More information

3.1 More on model selection

3.1 More on model selection 3. More on Model selecion 3. Comparing models AIC, BIC, Adjused R squared. 3. Over Fiing problem. 3.3 Sample spliing. 3. More on model selecion crieria Ofen afer model fiing you are lef wih a handful of

More information

Chemotaxis for Controller Tuning

Chemotaxis for Controller Tuning Chemoaxis for Conroller Tuning Kauko Leiviskä*, Iiris Joensuu** *Universi of Oulu, Conrol Engineering Laboraor P.O.Box 43, FIN-94 Oulun liopiso, Finland Phone: +358-8-55346, Fax: +358-8-55334 email:kauko.leiviska@oulu.fi

More information

Optimal Path Planning for Flexible Redundant Robot Manipulators

Optimal Path Planning for Flexible Redundant Robot Manipulators 25 WSEAS In. Conf. on DYNAMICAL SYSEMS and CONROL, Venice, Ialy, November 2-4, 25 (pp363-368) Opimal Pah Planning for Flexible Redundan Robo Manipulaors H. HOMAEI, M. KESHMIRI Deparmen of Mechanical Engineering

More information

WEEK-3 Recitation PHYS 131. of the projectile s velocity remains constant throughout the motion, since the acceleration a x

WEEK-3 Recitation PHYS 131. of the projectile s velocity remains constant throughout the motion, since the acceleration a x WEEK-3 Reciaion PHYS 131 Ch. 3: FOC 1, 3, 4, 6, 14. Problems 9, 37, 41 & 71 and Ch. 4: FOC 1, 3, 5, 8. Problems 3, 5 & 16. Feb 8, 018 Ch. 3: FOC 1, 3, 4, 6, 14. 1. (a) The horizonal componen of he projecile

More information

Shiva Akhtarian MSc Student, Department of Computer Engineering and Information Technology, Payame Noor University, Iran

Shiva Akhtarian MSc Student, Department of Computer Engineering and Information Technology, Payame Noor University, Iran Curren Trends in Technology and Science ISSN : 79-055 8hSASTech 04 Symposium on Advances in Science & Technology-Commission-IV Mashhad, Iran A New for Sofware Reliabiliy Evaluaion Based on NHPP wih Imperfec

More information

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

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

More information

Kinematics and kinematic functions

Kinematics and kinematic functions Kinemaics and kinemaic funcions Kinemaics deals wih he sudy of four funcions (called kinemaic funcions or KFs) ha mahemaically ransform join variables ino caresian variables and vice versa Direc Posiion

More information

Overview. COMP14112: Artificial Intelligence Fundamentals. Lecture 0 Very Brief Overview. Structure of this course

Overview. COMP14112: Artificial Intelligence Fundamentals. Lecture 0 Very Brief Overview. Structure of this course OMP: Arificial Inelligence Fundamenals Lecure 0 Very Brief Overview Lecurer: Email: Xiao-Jun Zeng x.zeng@mancheser.ac.uk Overview This course will focus mainly on probabilisic mehods in AI We shall presen

More information

Inventory Control of Perishable Items in a Two-Echelon Supply Chain

Inventory Control of Perishable Items in a Two-Echelon Supply Chain Journal of Indusrial Engineering, Universiy of ehran, Special Issue,, PP. 69-77 69 Invenory Conrol of Perishable Iems in a wo-echelon Supply Chain Fariborz Jolai *, Elmira Gheisariha and Farnaz Nojavan

More information

Notes on Kalman Filtering

Notes on Kalman Filtering Noes on Kalman Filering Brian Borchers and Rick Aser November 7, Inroducion Daa Assimilaion is he problem of merging model predicions wih acual measuremens of a sysem o produce an opimal esimae of he curren

More information

Lecture 4 Kinetics of a particle Part 3: Impulse and Momentum

Lecture 4 Kinetics of a particle Part 3: Impulse and Momentum MEE Engineering Mechanics II Lecure 4 Lecure 4 Kineics of a paricle Par 3: Impulse and Momenum Linear impulse and momenum Saring from he equaion of moion for a paricle of mass m which is subjeced o an

More information

Swarm. Particle Swarm Optimization. Swarm. Swarm. Swarm. Swarm

Swarm. Particle Swarm Optimization. Swarm. Swarm. Swarm. Swarm Paricle Opimizaion Krzyszof Troanowski IPI PAN & UKSW, Warszawa maa r Inroducion PSO Precursors In 986 I made a compuer model of coordinaed animal moion such as bird flocks and fish schools I was based

More information

Speaker Adaptation Techniques For Continuous Speech Using Medium and Small Adaptation Data Sets. Constantinos Boulis

Speaker Adaptation Techniques For Continuous Speech Using Medium and Small Adaptation Data Sets. Constantinos Boulis Speaker Adapaion Techniques For Coninuous Speech Using Medium and Small Adapaion Daa Ses Consaninos Boulis Ouline of he Presenaion Inroducion o he speaker adapaion problem Maximum Likelihood Sochasic Transformaions

More information

10. State Space Methods

10. State Space Methods . Sae Space Mehods. Inroducion Sae space modelling was briefly inroduced in chaper. Here more coverage is provided of sae space mehods before some of heir uses in conrol sysem design are covered in he

More information

arxiv: v1 [cs.dc] 8 Mar 2012

arxiv: v1 [cs.dc] 8 Mar 2012 Consensus on Moving Neighborhood Model of Peerson Graph arxiv:.9v [cs.dc] 8 Mar Hannah Arend and Jorgensen Jos Absrac In his paper, we sudy he consensus problem of muliple agens on a kind of famous graph,

More information