FEEDFORWARD NEURAL NETWORK COMBINED WITH NUMERICAL INTEGRATOR STRUCTURE FOR AN INVERTED PENDULUM MODELING AND SIMULATION.
|
|
- Katherine Hood
- 6 years ago
- Views:
Transcription
1 Anais d 12 O Encntr de Iniciaçã Científica e Pós-Graduaçã d ITA XII ENCITA / 26 Institut Tecnlógic de Aernáutica Sã Jsé ds Camps SP Brasil Outubr 16 a FEEDFORWARD NEURAL NETWORK COMBINED WITH NUMERICAL INTEGRATOR STRUCTURE FOR AN INVERTED PENDULUM MODELING AND SIMULATION. Daniel van Berghem Mtta Divisã de Ciência da Cmputaçã Institut Tecnlógic da Aernáutica - ITA Pç. Marechal Eduard Gmes 5 Vila das Acácias CEP Sã Jsé ds Camps SP Brasil dvbmtta@bl.cm.br Palma Maria Silva Rcha Departament de Engenharia Elétrica Universidade Estadual Paulista - UNESP Av. Dr. Aribert Pereira da Cunha 333 Pedregulh CEP Guaratinguetá SP Brasil. palma@feg.unesp.br Paul Marcel Tasinaff Divisã de Ciência da Cmputaçã Institut Tecnlógic da Aernáutica - ITA Pç. Marechal Eduard Gmes 5 Vila das Acácias CEP Sã Jsé ds Camps SP Brasil tasinaf@ita.br Abstract: This article aims t make the simulatin f nn-linearized inverted pendulum mdels using the set neural artificial netwrk cmbined with high accuracy numerical integratin. T achieve this gal we used a feedfward neural netwrk and the gradient and kalman algrithms t train this net. Using this algrithms we addressed the questin f cmputacinal csts t train and simulate these mdels. These csts includes CPU time t prcess the neural netwrk training and alcatin f memry t dispse data. As the gradient training methd in the neural netwrk gives the derivative functin f the system ( y`= f (yt) ) we used numeric integratrs t btain the utput f the system. S a furth rder Runge-Kutta and an Adams-Bashfrth integratr were matched with the utput f the neural netwrk ( y` ) s that we culd have y. Then it was pssible t analyze graphically the perfrmance f these integratin algrithms cmpared with the true utput f the physic dynamic system given by the slutin f the mdel`s analytic differential equatin. Key wrds: FeedFrward Neural Netwrk Inverted Pendulum Mdel Numerical Integratin nn-linear system Kalman Filter learning gradient methd. 1. Intrdutin Neural netwrks have been used widely fr identificatin f nn-linear dynamic systems (Carrara 1997; Chen et al 1992 and Hunt et al 1992). A feedfrward neural netwrk can be used in the structure f an rdinary differencial equatin numerical integratr algrithm t apprximate and t replace the derivative functin f the rdinary differential equatins dynamic system mathematical mdel a discrete mdel is btained which has quite advantageus characteristics (Ris Net 21 and Wang and Lin 1998). With this apprach the difficulty with t many inputs in the training f the neural netwrk is alleviated since it is nly necessary t learn an algebraic and static functin and the inputs are ccurrences f the state and cntrl variables in their envelpe f variatin. It is pssible fr example t train a feedfrward neural netwrk in instantaneus derivate methd with was prpsed by Wang and Lin in 1998 t represent nn-linear dynamic systems. In this paper we used a feedfward neural netwrk and instantaneus derivate methd t represent a nn-linear and a linear inverted pendulum mdel using Gradient and Kalman algrithms t train f the net. All this training used the sftware implemented by Tasinaff in 24. This sftware enabled us t vary several parameters f the net cnfiguratin such as integratin step and s we culd bserve their influence in the MEQ (Average Squared Errr) and hw they differently affect linear and nn-linear mdel training. Using this sftware we addressed the questin f cmputatinal csts t train and simulate these mdels. These csts includes CPU time t prcess the neural netwrk training and lcatin f memry t dispse data. Thugh this paper is fcused in testing the prcedure in an applicatin f interest it is rganized t guarantee a cmprehensive presentatin fr the benefit f thse nt familiar with the tested prcedure. In what fllws Sectin 2 presents the fundaments f feedfrward neural netwrks. Sectin 3 presents the cncepts f numerical integratr f discrete frward mdel and neural numerical integratr. In sectin 4 the simulatin and the results f the applicatin t the prblem f mdeling the inverted pendulum are presented. In the last sectin sme cnclusins and suggestins t future researches are drawn.
2 Anais d XII ENCITA 26 ITA Outubr Fundamentals: FeedFrward Neural Netwrks. It is nt easy t find ut a unique definitin abut what is an artificial neural netwrk (Zurada 22). In very general terms artificial neural netwrks are cmputatin mdels made up f artificial cmputatinally mdeled neurns that result frm explring analgies with bilgical neural netwrks f life rganisms. The feedfrward multiplayer perceptrn is a neural netwrk where Perceptrn type f artificial neurns (Braga et al 2) (Figure 1) are rganized in layers cnnected in a feedfrward architecture. The Perceptrn neurn is a classifier where inputs (xj s) are initially classified with respect t a hyperplane characterized by the weights (wij s) and a bias (bi) in series with a secnd nn linear classificatin dne by an activatin functin (fi(.)) which limits the classificatin t be inside asympttic nrmalized limits f (1) in the case f the Lgsig activatin r (-1+1) in the case f the Tansig activatin. nk-1 k k k k-1 k x i = f w ij x j b i p/ i = n k j= 1 (1) Figure 1 The Perceptrn Neurn. The Feedfrward neural netwrk is made up f clumns f Perceptrns (f s) and s the utput vectr x k f a layer k is illustrate and simplificatin as in figure 2. Figure 2 Input/Output f a General Hidden Layer k. 3. Numerical Integratr Discrete Frward Mdel and Neural Numerical Integratr Cnsider a dynamic system with a mathematical mdel given by a set f rdinary differential equatins: x = f(x u) (2) where x Є R n is the state vectr; u Є R m is the cntrl vectr; f( x u) is the derivative functin cming frm physical laws gverning the dynamic system. Cnsider nw an rdinary differential equatin (ODE) numerical integratr (Ster et al 198) t get a discrete apprximatin f the system f Eq.(3): x = x( t + t) f ( x( t) x( t t)... x( t n t); u( t)... u( t n t); t) (3) n
3 Anais d XII ENCITA 26 ITA Outubr f ( x( t) x( t t)... x( t n t); u( t)... u( t n t); t) is the functin that results frm using where n evaluatins f the derivative functin f Eq. (2) in the numerical integratr algrithm; n is related t the rder f the apprximatin; if its value is greater than zer ne has the situatin where a finite difference type f integratr is used (fr example an Adams-Bashfrth methd); if it is zer a single step type f integratr is used (fr example a Runge- Kutta methd); and t the step size is assumed sufficiently small t assure u(t) cnstant alng the discretizatin interval. The numerical integratr in Eq.(3) can be used recursively as an apprximate discrete predictive mdel f the dynamic system f Eq.(2) in internal mdel cntrl schemes. The errr in each step can be cntrlled by varying step size and r the rder f numerical integratr; and the resulting numerical algrithm can be prcessed in parallel fr each cmpnent f the state f the dynamic system. In the dynamic system f Eq. (2) the derivative functin in the ODE mathematical mdel is an algebraic functin that can be apprximated by a feedfrward neural netwrk: x = f ( x u) fˆ ( x u wˆ ) (4) f ( x u w ˆ ) is t represent the neural netwrk trained; and ŵ the neural netwrk learned weights. Cnsider nw this neural apprximatin f the derivative functin in the structure f an rdinary differential equatin (ODE) numerical integratr (Ster et al 198) t get a discrete apprximatin f the system f Eq.(2) : where ˆ x( t + t) fˆ ( x( t) x( t t)... x( t n t); u( t)... u( t n t); t; wˆ ) (5) n The neural numerical integratr in Eq.(5) is an apprximatin f the ODE numerical integratr f Eq. (3) and thus an apprximate discrete predictive mdel f the dynamic system f Eq.(2) which can be used as an internal mdel in cntrl schemes. 4. Simulatin and Results As shwn belw in figure 3 ur mdel cnsists f an inverted pendulum. We mdeled it using the appraches f a nn-linear mdel. The table 1 gives the nminal values used fr mdeling the bdy cnsiderate. Figure 3 Inverted pendulum. Table 1 - Inverted pendulum parameters. The nn-linear mdel can be described by the fllwing equatins where x and theta are as abve. M + m x + bx + ml θ θ ml θ θ = F 2 ( I + ml ) θ + mgl sinθ = mlx csθ (6) 2 ( ) cs sin
4 Anais d XII ENCITA 26 ITA Outubr Our bjective with this paper is divided in tw tasks: train a Neural Netwrk (NN) t learn the differential equatin in term f state variables that fits the nn-linear mdel f the inverted pendulum and after simulate the time respnse f the mdel t different initial cnditins. T train the NN we adjusted the number f training patterns given t the net and the learning rate s that we culd have a mdel gd enugh and minimize the csts in terms f prcessing. Besides we varied: x frm -5 t 5 meters; dx/dt frm -14 t 14 m/s; theta frm pi/12 t pi/12; d(theta)/dt: frm -1 t 1 rad/s F ( external excitatin f the system ) frm -1 t 1 Newtns We attempt t reach MEQ in the training f the NN abut 1^-7.Infrtunately the minimum MEQ we gt in nnlinear case was *1^-5. We used 35 patterns pints generated randmly within the state-space described abve. In additin we used tw algrithms t train the net : Kalman Filter and Gradient. The functining f this algrithms aren t in the scpe f ur wrk. After setting the main parameters f the simulatin as the step f integratin initial cnditin f state variables e time interval f simulatin we culd get sme results as shwn belw. Representaça Neural de Sistema Dinamic Na-Linear.2.1 c/ MEQ=1.642e-6 State Varible State Varible State Varible Original Furth Order.5Runge-Kutta 1 Neural Furth Nrmalized Order Runge-Kutta Time.6 Neural Furth Order Adams-Bashfrth State Varible Figure 4 Graphics shwing the evlutin in time interval [1] f state variables f the nn-linear mdel. As shwn in Figure 4 the time interval f simulatin was reduced frm ( as the initial time f simulatin ) until 1 ( end pint f simulatin ). It s easy t see that this simulatin was much mre successful than the first ne. It s due t the capability f apprximatin f the nn-linear mdel. Thus [1] is a reliable interval t this mdel. The prpagatin MEQ was drastically decreased with this change in simulatin interval. Representaça Neural de Sistema Dinamic Na-Linear.2 c/ MEQ=1.2812e-5 State Varible State Varible State Varible Original Furth Order Runge-Kutta Neural.4 Furth Order Runge-Kutta Neural Furth Order Adams-Bashfrth.2 State Varible Figure 5 Graphics shwing the evlutin in time interval [3] f state variables f the nn-linear mdel.
5 Anais d XII ENCITA 26 ITA Outubr In Figure 5 it s clear that we have reached a gd result with the nn-linear apprach with the new interval [ 3] secnd. This cnfirms the imprtance f the nn-linearity in the inverted pendulum mdel that itself can be applied besides the linear cnfinatin. It s very imprtant t ntice that the step f integratin cnsiderably affects the precisin f the neural netwrk estimatin. Nevertheless decreasing the step f integratin means an increase in the csts f cmputer prcessing time. 5. Cnclusins and suggestins t future researches This article reached the ur aims namely: the limitatins f a mathematical mdel has t represent a physical ne; the implementatins and simulatins f a nn-linear general mdel; the ptentialities and limitatins f Feedfrward neural netwrk with a single internal layer t represent and learn a nn-linear mdel and the influence f NN parameters and settings in the its perfrmance and thrughput. S we hpe we have cntributed t the understanding f FeedFrwad NN and its implementatin and as suggestin it wuld be very interesting t develp this prblem using a RBF Neural Netwrk architecture. In this case the RBF is a gd estimatr f linear parameters and pssibly the results f the linear mdel estimatin wuld be better. Afterwards it wuld be interesting as well as t implement a FeedFrward NN with mre than ne internal layer t see hw it wuld behave in terms f nn linear estimatin and cmputatinal csts. 6. References Braga A. P. Ludermir T. B. Carvalh A. C. P. L. F 2 Redes Neurais Artificiais Teria e Aplicações. LTC. Carrara V Redes Neurais Aplicadas a Cntrle de Atitude de Satélites cm Gemetria Variável Tese de Dutrad INPE Sã Jsé ds Camps. Chen S. Billings S. A Neural Netwrks fr Nnlinear Dynamic System Mdeling and Identificatin Int. J. Cntrl Vl. 56 N. 2 pp Hunt K. J. Sbarbar D. Zbikwski R. Gawthrp P. J Neural Netwrks fr Cntrl Systems A Survey Autmatica Vl. 28 N. 6 pp Ris Net A. 21 Dynamic Systems Numerical Integratrs in Neural Cntrl Schemes V Cngress Brasileir de Redes Neurais. Ster J. and Bulirsch R. 198 Intrductin t Numerical Analisys Springer Verlag N.Y.. Tassinaf M. 24 Estruturas de integraçã neural Feedfward testadas em prblemas de cntrle preditiv Tese de Dutrad INPE Sã Jsé ds Camps. Wang Y.-J. Lin C.-T Runge-Kutta Neural Netwrk fr Identificatin f Dynamical Systems in High Accuracy IEEE Transactins On Neural Netwrks Vl. 9 N. 2 pp Zurada J. M Intrductin t Artificial Neural System. West Pub. C.
the results to larger systems due to prop'erties of the projection algorithm. First, the number of hidden nodes must
M.E. Aggune, M.J. Dambrg, M.A. El-Sharkawi, R.J. Marks II and L.E. Atlas, "Dynamic and static security assessment f pwer systems using artificial neural netwrks", Prceedings f the NSF Wrkshp n Applicatins
More informationBootstrap Method > # Purpose: understand how bootstrap method works > obs=c(11.96, 5.03, 67.40, 16.07, 31.50, 7.73, 11.10, 22.38) > n=length(obs) >
Btstrap Methd > # Purpse: understand hw btstrap methd wrks > bs=c(11.96, 5.03, 67.40, 16.07, 31.50, 7.73, 11.10, 22.38) > n=length(bs) > mean(bs) [1] 21.64625 > # estimate f lambda > lambda = 1/mean(bs);
More informationComputational modeling techniques
Cmputatinal mdeling techniques Lecture 4: Mdel checing fr ODE mdels In Petre Department f IT, Åb Aademi http://www.users.ab.fi/ipetre/cmpmd/ Cntent Stichimetric matrix Calculating the mass cnservatin relatins
More informationMINIMIZATION OF ACTUATOR REPOSITIONING USING NEURAL NETWORKS WITH APPLICATION IN NONLINEAR HVAC 1 SYSTEMS
MINIMIZATION OF ACTUATOR REPOSITIONING USING NEURAL NETWORKS WITH APPLICATION IN NONLINEAR HVAC SYSTEMS M. J. Yazdanpanah *, E. Semsar, C. Lucas * yazdan@ut.ac.ir, semsar@chamran.ut.ac.ir, lucas@ipm.ir
More informationA Scalable Recurrent Neural Network Framework for Model-free
A Scalable Recurrent Neural Netwrk Framewrk fr Mdel-free POMDPs April 3, 2007 Zhenzhen Liu, Itamar Elhanany Machine Intelligence Lab Department f Electrical and Cmputer Engineering The University f Tennessee
More informationMODULE FOUR. This module addresses functions. SC Academic Elementary Algebra Standards:
MODULE FOUR This mdule addresses functins SC Academic Standards: EA-3.1 Classify a relatinship as being either a functin r nt a functin when given data as a table, set f rdered pairs, r graph. EA-3.2 Use
More informationinitially lcated away frm the data set never win the cmpetitin, resulting in a nnptimal nal cdebk, [2] [3] [4] and [5]. Khnen's Self Organizing Featur
Cdewrd Distributin fr Frequency Sensitive Cmpetitive Learning with One Dimensinal Input Data Aristides S. Galanpuls and Stanley C. Ahalt Department f Electrical Engineering The Ohi State University Abstract
More informationDesign and Simulation of Dc-Dc Voltage Converters Using Matlab/Simulink
American Jurnal f Engineering Research (AJER) 016 American Jurnal f Engineering Research (AJER) e-issn: 30-0847 p-issn : 30-0936 Vlume-5, Issue-, pp-9-36 www.ajer.rg Research Paper Open Access Design and
More informationComputational modeling techniques
Cmputatinal mdeling techniques Lecture 2: Mdeling change. In Petre Department f IT, Åb Akademi http://users.ab.fi/ipetre/cmpmd/ Cntent f the lecture Basic paradigm f mdeling change Examples Linear dynamical
More informationEnhancing Performance of MLP/RBF Neural Classifiers via an Multivariate Data Distribution Scheme
Enhancing Perfrmance f / Neural Classifiers via an Multivariate Data Distributin Scheme Halis Altun, Gökhan Gelen Nigde University, Electrical and Electrnics Engineering Department Nigde, Turkey haltun@nigde.edu.tr
More informationIntroduction: A Generalized approach for computing the trajectories associated with the Newtonian N Body Problem
A Generalized apprach fr cmputing the trajectries assciated with the Newtnian N Bdy Prblem AbuBar Mehmd, Syed Umer Abbas Shah and Ghulam Shabbir Faculty f Engineering Sciences, GIK Institute f Engineering
More information1996 Engineering Systems Design and Analysis Conference, Montpellier, France, July 1-4, 1996, Vol. 7, pp
THE POWER AND LIMIT OF NEURAL NETWORKS T. Y. Lin Department f Mathematics and Cmputer Science San Jse State University San Jse, Califrnia 959-003 tylin@cs.ssu.edu and Bereley Initiative in Sft Cmputing*
More informationChE 471: LECTURE 4 Fall 2003
ChE 47: LECTURE 4 Fall 003 IDEL RECTORS One f the key gals f chemical reactin engineering is t quantify the relatinship between prductin rate, reactr size, reactin kinetics and selected perating cnditins.
More informationLeast Squares Optimal Filtering with Multirate Observations
Prc. 36th Asilmar Cnf. n Signals, Systems, and Cmputers, Pacific Grve, CA, Nvember 2002 Least Squares Optimal Filtering with Multirate Observatins Charles W. herrien and Anthny H. Hawes Department f Electrical
More informationSIZE BIAS IN LINE TRANSECT SAMPLING: A FIELD TEST. Mark C. Otto Statistics Research Division, Bureau of the Census Washington, D.C , U.S.A.
SIZE BIAS IN LINE TRANSECT SAMPLING: A FIELD TEST Mark C. Ott Statistics Research Divisin, Bureau f the Census Washingtn, D.C. 20233, U.S.A. and Kenneth H. Pllck Department f Statistics, Nrth Carlina State
More informationAP Physics Kinematic Wrap Up
AP Physics Kinematic Wrap Up S what d yu need t knw abut this mtin in tw-dimensin stuff t get a gd scre n the ld AP Physics Test? First ff, here are the equatins that yu ll have t wrk with: v v at x x
More informationENSC Discrete Time Systems. Project Outline. Semester
ENSC 49 - iscrete Time Systems Prject Outline Semester 006-1. Objectives The gal f the prject is t design a channel fading simulatr. Upn successful cmpletin f the prject, yu will reinfrce yur understanding
More informationCAUSAL INFERENCE. Technical Track Session I. Phillippe Leite. The World Bank
CAUSAL INFERENCE Technical Track Sessin I Phillippe Leite The Wrld Bank These slides were develped by Christel Vermeersch and mdified by Phillippe Leite fr the purpse f this wrkshp Plicy questins are causal
More informationMODULE 1. e x + c. [You can t separate a demominator, but you can divide a single denominator into each numerator term] a + b a(a + b)+1 = a + b
. REVIEW OF SOME BASIC ALGEBRA MODULE () Slving Equatins Yu shuld be able t slve fr x: a + b = c a d + e x + c and get x = e(ba +) b(c a) d(ba +) c Cmmn mistakes and strategies:. a b + c a b + a c, but
More informationDetermining the Accuracy of Modal Parameter Estimation Methods
Determining the Accuracy f Mdal Parameter Estimatin Methds by Michael Lee Ph.D., P.E. & Mar Richardsn Ph.D. Structural Measurement Systems Milpitas, CA Abstract The mst cmmn type f mdal testing system
More informationSupport-Vector Machines
Supprt-Vectr Machines Intrductin Supprt vectr machine is a linear machine with sme very nice prperties. Haykin chapter 6. See Alpaydin chapter 13 fr similar cntent. Nte: Part f this lecture drew material
More informationDead-beat controller design
J. Hetthéssy, A. Barta, R. Bars: Dead beat cntrller design Nvember, 4 Dead-beat cntrller design In sampled data cntrl systems the cntrller is realised by an intelligent device, typically by a PLC (Prgrammable
More informationA Matrix Representation of Panel Data
web Extensin 6 Appendix 6.A A Matrix Representatin f Panel Data Panel data mdels cme in tw brad varieties, distinct intercept DGPs and errr cmpnent DGPs. his appendix presents matrix algebra representatins
More informationSlide04 (supplemental) Haykin Chapter 4 (both 2nd and 3rd ed): Multi-Layer Perceptrons
Slide04 supplemental) Haykin Chapter 4 bth 2nd and 3rd ed): Multi-Layer Perceptrns CPSC 636-600 Instructr: Ynsuck Che Heuristic fr Making Backprp Perfrm Better 1. Sequential vs. batch update: fr large
More informationModelling of Clock Behaviour. Don Percival. Applied Physics Laboratory University of Washington Seattle, Washington, USA
Mdelling f Clck Behaviur Dn Percival Applied Physics Labratry University f Washingtn Seattle, Washingtn, USA verheads and paper fr talk available at http://faculty.washingtn.edu/dbp/talks.html 1 Overview
More informationLinearization of the Output of a Wheatstone Bridge for Single Active Sensor. Madhu Mohan N., Geetha T., Sankaran P. and Jagadeesh Kumar V.
Linearizatin f the Output f a Wheatstne Bridge fr Single Active Sensr Madhu Mhan N., Geetha T., Sankaran P. and Jagadeesh Kumar V. Dept. f Electrical Engineering, Indian Institute f Technlgy Madras, Chennai
More informationRevision: August 19, E Main Suite D Pullman, WA (509) Voice and Fax
.7.4: Direct frequency dmain circuit analysis Revisin: August 9, 00 5 E Main Suite D Pullman, WA 9963 (509) 334 6306 ice and Fax Overview n chapter.7., we determined the steadystate respnse f electrical
More informationOn Huntsberger Type Shrinkage Estimator for the Mean of Normal Distribution ABSTRACT INTRODUCTION
Malaysian Jurnal f Mathematical Sciences 4(): 7-4 () On Huntsberger Type Shrinkage Estimatr fr the Mean f Nrmal Distributin Department f Mathematical and Physical Sciences, University f Nizwa, Sultanate
More informationKinetic Model Completeness
5.68J/10.652J Spring 2003 Lecture Ntes Tuesday April 15, 2003 Kinetic Mdel Cmpleteness We say a chemical kinetic mdel is cmplete fr a particular reactin cnditin when it cntains all the species and reactins
More informationCHAPTER 24: INFERENCE IN REGRESSION. Chapter 24: Make inferences about the population from which the sample data came.
MATH 1342 Ch. 24 April 25 and 27, 2013 Page 1 f 5 CHAPTER 24: INFERENCE IN REGRESSION Chapters 4 and 5: Relatinships between tw quantitative variables. Be able t Make a graph (scatterplt) Summarize the
More informationResampling Methods. Cross-validation, Bootstrapping. Marek Petrik 2/21/2017
Resampling Methds Crss-validatin, Btstrapping Marek Petrik 2/21/2017 Sme f the figures in this presentatin are taken frm An Intrductin t Statistical Learning, with applicatins in R (Springer, 2013) with
More informationAnalysis on the Stability of Reservoir Soil Slope Based on Fuzzy Artificial Neural Network
Research Jurnal f Applied Sciences, Engineering and Technlgy 5(2): 465-469, 2013 ISSN: 2040-7459; E-ISSN: 2040-7467 Maxwell Scientific Organizatin, 2013 Submitted: May 08, 2012 Accepted: May 29, 2012 Published:
More informationYield-driven em optimization using space mapping-based neuromodels
Institut Tecnlógic y de Estudis Superires de Occidente Repsitri Institucinal del ITESO rei.ites.mx Departament de Electrónica, Sistemas e Inrmática DESI - Artículs y pnencias cn arbitraje 2-9 Yield-driven
More informationChapter 3: Cluster Analysis
Chapter 3: Cluster Analysis } 3.1 Basic Cncepts f Clustering 3.1.1 Cluster Analysis 3.1. Clustering Categries } 3. Partitining Methds 3..1 The principle 3.. K-Means Methd 3..3 K-Medids Methd 3..4 CLARA
More informationCOMP 551 Applied Machine Learning Lecture 5: Generative models for linear classification
COMP 551 Applied Machine Learning Lecture 5: Generative mdels fr linear classificatin Instructr: Herke van Hf (herke.vanhf@mail.mcgill.ca) Slides mstly by: Jelle Pineau Class web page: www.cs.mcgill.ca/~hvanh2/cmp551
More informationChapter 3 Kinematics in Two Dimensions; Vectors
Chapter 3 Kinematics in Tw Dimensins; Vectrs Vectrs and Scalars Additin f Vectrs Graphical Methds (One and Tw- Dimensin) Multiplicatin f a Vectr b a Scalar Subtractin f Vectrs Graphical Methds Adding Vectrs
More informationAdmissibility Conditions and Asymptotic Behavior of Strongly Regular Graphs
Admissibility Cnditins and Asympttic Behavir f Strngly Regular Graphs VASCO MOÇO MANO Department f Mathematics University f Prt Oprt PORTUGAL vascmcman@gmailcm LUÍS ANTÓNIO DE ALMEIDA VIEIRA Department
More informationArtificial Neural Networks MLP, Backpropagation
Artificial Neural Netwrks MLP, Backprpagatin 01001110 01100101 01110101 01110010 01101111 01101110 01101111 01110110 01100001 00100000 01110011 01101011 01110101 01110000 01101001 01101110 01100001 00100000
More informationFall 2013 Physics 172 Recitation 3 Momentum and Springs
Fall 03 Physics 7 Recitatin 3 Mmentum and Springs Purpse: The purpse f this recitatin is t give yu experience wrking with mmentum and the mmentum update frmula. Readings: Chapter.3-.5 Learning Objectives:.3.
More informationKinetics of Particles. Chapter 3
Kinetics f Particles Chapter 3 1 Kinetics f Particles It is the study f the relatins existing between the frces acting n bdy, the mass f the bdy, and the mtin f the bdy. It is the study f the relatin between
More informationFlipping Physics Lecture Notes: Simple Harmonic Motion Introduction via a Horizontal Mass-Spring System
Flipping Physics Lecture Ntes: Simple Harmnic Mtin Intrductin via a Hrizntal Mass-Spring System A Hrizntal Mass-Spring System is where a mass is attached t a spring, riented hrizntally, and then placed
More informationComputational modeling techniques
Cmputatinal mdeling techniques Lecture 11: Mdeling with systems f ODEs In Petre Department f IT, Ab Akademi http://www.users.ab.fi/ipetre/cmpmd/ Mdeling with differential equatins Mdeling strategy Fcus
More informationSPH3U1 Lesson 06 Kinematics
PROJECTILE MOTION LEARNING GOALS Students will: Describe the mtin f an bject thrwn at arbitrary angles thrugh the air. Describe the hrizntal and vertical mtins f a prjectile. Slve prjectile mtin prblems.
More informationCambridge Assessment International Education Cambridge Ordinary Level. Published
Cambridge Assessment Internatinal Educatin Cambridge Ordinary Level ADDITIONAL MATHEMATICS 4037/1 Paper 1 Octber/Nvember 017 MARK SCHEME Maximum Mark: 80 Published This mark scheme is published as an aid
More information[COLLEGE ALGEBRA EXAM I REVIEW TOPICS] ( u s e t h i s t o m a k e s u r e y o u a r e r e a d y )
(Abut the final) [COLLEGE ALGEBRA EXAM I REVIEW TOPICS] ( u s e t h i s t m a k e s u r e y u a r e r e a d y ) The department writes the final exam s I dn't really knw what's n it and I can't very well
More informationCS 477/677 Analysis of Algorithms Fall 2007 Dr. George Bebis Course Project Due Date: 11/29/2007
CS 477/677 Analysis f Algrithms Fall 2007 Dr. Gerge Bebis Curse Prject Due Date: 11/29/2007 Part1: Cmparisn f Srting Algrithms (70% f the prject grade) The bjective f the first part f the assignment is
More informationECEN 4872/5827 Lecture Notes
ECEN 4872/5827 Lecture Ntes Lecture #5 Objectives fr lecture #5: 1. Analysis f precisin current reference 2. Appraches fr evaluating tlerances 3. Temperature Cefficients evaluatin technique 4. Fundamentals
More informationk-nearest Neighbor How to choose k Average of k points more reliable when: Large k: noise in attributes +o o noise in class labels
Mtivating Example Memry-Based Learning Instance-Based Learning K-earest eighbr Inductive Assumptin Similar inputs map t similar utputs If nt true => learning is impssible If true => learning reduces t
More informationCollocation Map for Overcoming Data Sparseness
Cllcatin Map fr Overcming Data Sparseness Mnj Kim, Yung S. Han, and Key-Sun Chi Department f Cmputer Science Krea Advanced Institute f Science and Technlgy Taejn, 305-701, Krea mj0712~eve.kaist.ac.kr,
More informationSimple Linear Regression (single variable)
Simple Linear Regressin (single variable) Intrductin t Machine Learning Marek Petrik January 31, 2017 Sme f the figures in this presentatin are taken frm An Intrductin t Statistical Learning, with applicatins
More informationNUROP CONGRESS PAPER CHINESE PINYIN TO CHINESE CHARACTER CONVERSION
NUROP Chinese Pinyin T Chinese Character Cnversin NUROP CONGRESS PAPER CHINESE PINYIN TO CHINESE CHARACTER CONVERSION CHIA LI SHI 1 AND LUA KIM TENG 2 Schl f Cmputing, Natinal University f Singapre 3 Science
More informationFlipping Physics Lecture Notes: Simple Harmonic Motion Introduction via a Horizontal Mass-Spring System
Flipping Physics Lecture Ntes: Simple Harmnic Mtin Intrductin via a Hrizntal Mass-Spring System A Hrizntal Mass-Spring System is where a mass is attached t a spring, riented hrizntally, and then placed
More informationLead/Lag Compensator Frequency Domain Properties and Design Methods
Lectures 6 and 7 Lead/Lag Cmpensatr Frequency Dmain Prperties and Design Methds Definitin Cnsider the cmpensatr (ie cntrller Fr, it is called a lag cmpensatr s K Fr s, it is called a lead cmpensatr Ntatin
More informationROUNDING ERRORS IN BEAM-TRACKING CALCULATIONS
Particle Acceleratrs, 1986, Vl. 19, pp. 99-105 0031-2460/86/1904-0099/$15.00/0 1986 Grdn and Breach, Science Publishers, S.A. Printed in the United States f America ROUNDING ERRORS IN BEAM-TRACKING CALCULATIONS
More informationTechnical Bulletin. Generation Interconnection Procedures. Revisions to Cluster 4, Phase 1 Study Methodology
Technical Bulletin Generatin Intercnnectin Prcedures Revisins t Cluster 4, Phase 1 Study Methdlgy Release Date: Octber 20, 2011 (Finalizatin f the Draft Technical Bulletin released n September 19, 2011)
More informationMultiple Source Multiple. using Network Coding
Multiple Surce Multiple Destinatin Tplgy Inference using Netwrk Cding Pegah Sattari EECS, UC Irvine Jint wrk with Athina Markpulu, at UCI, Christina Fraguli, at EPFL, Lausanne Outline Netwrk Tmgraphy Gal,
More informationSection I5: Feedback in Operational Amplifiers
Sectin I5: eedback in Operatinal mplifiers s discussed earlier, practical p-amps hae a high gain under dc (zer frequency) cnditins and the gain decreases as frequency increases. This frequency dependence
More informationFree Vibrations of Catenary Risers with Internal Fluid
Prceeding Series f the Brazilian Sciety f Applied and Cmputatinal Mathematics, Vl. 4, N. 1, 216. Trabalh apresentad n DINCON, Natal - RN, 215. Prceeding Series f the Brazilian Sciety f Cmputatinal and
More informationand the Doppler frequency rate f R , can be related to the coefficients of this polynomial. The relationships are:
Algrithm fr Estimating R and R - (David Sandwell, SIO, August 4, 2006) Azimith cmpressin invlves the alignment f successive eches t be fcused n a pint target Let s be the slw time alng the satellite track
More informationI. Analytical Potential and Field of a Uniform Rod. V E d. The definition of electric potential difference is
Length L>>a,b,c Phys 232 Lab 4 Ch 17 Electric Ptential Difference Materials: whitebards & pens, cmputers with VPythn, pwer supply & cables, multimeter, crkbard, thumbtacks, individual prbes and jined prbes,
More informationThe Research on Flux Linkage Characteristic Based on BP and RBF Neural Network for Switched Reluctance Motor
Prgress In Electrmagnetics Research M, Vl. 35, 5 6, 24 The Research n Flux Linkage Characteristic Based n BP and RBF Neural Netwrk fr Switched Reluctance Mtr Yan Cai, *, Siyuan Sun, Chenhui Wang, and Cha
More informationProtection of ungrounded systems using an advanced relay element
ENG 460 Prtectin f ungrunded systems using an advanced relay element A reprt submitted t the schl f Engineering and Energy, Murdch University in partial fulfilment f the requirements fr the degree f Bachelr
More informationPattern Recognition 2014 Support Vector Machines
Pattern Recgnitin 2014 Supprt Vectr Machines Ad Feelders Universiteit Utrecht Ad Feelders ( Universiteit Utrecht ) Pattern Recgnitin 1 / 55 Overview 1 Separable Case 2 Kernel Functins 3 Allwing Errrs (Sft
More informationmaking triangle (ie same reference angle) ). This is a standard form that will allow us all to have the X= y=
Intrductin t Vectrs I 21 Intrductin t Vectrs I 22 I. Determine the hrizntal and vertical cmpnents f the resultant vectr by cunting n the grid. X= y= J. Draw a mangle with hrizntal and vertical cmpnents
More informationCOMP 551 Applied Machine Learning Lecture 11: Support Vector Machines
COMP 551 Applied Machine Learning Lecture 11: Supprt Vectr Machines Instructr: (jpineau@cs.mcgill.ca) Class web page: www.cs.mcgill.ca/~jpineau/cmp551 Unless therwise nted, all material psted fr this curse
More informationRelationships Between Frequency, Capacitance, Inductance and Reactance.
P Physics Relatinships between f,, and. Relatinships Between Frequency, apacitance, nductance and Reactance. Purpse: T experimentally verify the relatinships between f, and. The data cllected will lead
More informationSections 15.1 to 15.12, 16.1 and 16.2 of the textbook (Robbins-Miller) cover the materials required for this topic.
Tpic : AC Fundamentals, Sinusidal Wavefrm, and Phasrs Sectins 5. t 5., 6. and 6. f the textbk (Rbbins-Miller) cver the materials required fr this tpic.. Wavefrms in electrical systems are current r vltage
More informationCOMP 551 Applied Machine Learning Lecture 4: Linear classification
COMP 551 Applied Machine Learning Lecture 4: Linear classificatin Instructr: Jelle Pineau (jpineau@cs.mcgill.ca) Class web page: www.cs.mcgill.ca/~jpineau/cmp551 Unless therwise nted, all material psted
More informationBeam vibrations: Discrete mass and stiffness models
Beam vibratins: Discrete mass and stiffness mdels Ana Cláudia Susa Neves ana.neves@tecnic.ulisba.pt Institut Superir Técnic, Universidade de Lisba, Prtugal May, 2015 Abstract In the present wrk the dynamic
More informationx 1 Outline IAML: Logistic Regression Decision Boundaries Example Data
Outline IAML: Lgistic Regressin Charles Suttn and Victr Lavrenk Schl f Infrmatics Semester Lgistic functin Lgistic regressin Learning lgistic regressin Optimizatin The pwer f nn-linear basis functins Least-squares
More informationOptimization Programming Problems For Control And Management Of Bacterial Disease With Two Stage Growth/Spread Among Plants
Internatinal Jurnal f Engineering Science Inventin ISSN (Online): 9 67, ISSN (Print): 9 676 www.ijesi.rg Vlume 5 Issue 8 ugust 06 PP.0-07 Optimizatin Prgramming Prblems Fr Cntrl nd Management Of Bacterial
More informationYou need to be able to define the following terms and answer basic questions about them:
CS440/ECE448 Sectin Q Fall 2017 Midterm Review Yu need t be able t define the fllwing terms and answer basic questins abut them: Intr t AI, agents and envirnments Pssible definitins f AI, prs and cns f
More informationNUMBERS, MATHEMATICS AND EQUATIONS
AUSTRALIAN CURRICULUM PHYSICS GETTING STARTED WITH PHYSICS NUMBERS, MATHEMATICS AND EQUATIONS An integral part t the understanding f ur physical wrld is the use f mathematical mdels which can be used t
More informationWe can see from the graph above that the intersection is, i.e., [ ).
MTH 111 Cllege Algebra Lecture Ntes July 2, 2014 Functin Arithmetic: With nt t much difficulty, we ntice that inputs f functins are numbers, and utputs f functins are numbers. S whatever we can d with
More informationThermodynamics Partial Outline of Topics
Thermdynamics Partial Outline f Tpics I. The secnd law f thermdynamics addresses the issue f spntaneity and invlves a functin called entrpy (S): If a prcess is spntaneus, then Suniverse > 0 (2 nd Law!)
More informationData Mining: Concepts and Techniques. Classification and Prediction. Chapter February 8, 2007 CSE-4412: Data Mining 1
Data Mining: Cncepts and Techniques Classificatin and Predictin Chapter 6.4-6 February 8, 2007 CSE-4412: Data Mining 1 Chapter 6 Classificatin and Predictin 1. What is classificatin? What is predictin?
More informationBuilding to Transformations on Coordinate Axis Grade 5: Geometry Graph points on the coordinate plane to solve real-world and mathematical problems.
Building t Transfrmatins n Crdinate Axis Grade 5: Gemetry Graph pints n the crdinate plane t slve real-wrld and mathematical prblems. 5.G.1. Use a pair f perpendicular number lines, called axes, t define
More informationAdmin. MDP Search Trees. Optimal Quantities. Reinforcement Learning
Admin Reinfrcement Learning Cntent adapted frm Berkeley CS188 MDP Search Trees Each MDP state prjects an expectimax-like search tree Optimal Quantities The value (utility) f a state s: V*(s) = expected
More informationCombining Dialectical Optimization and Gradient Descent Methods for Improving the Accuracy of Straight Line Segment Classifiers
Cmbining Dialectical Optimizatin and Gradient Descent Methds fr Imprving the Accuracy f Straight Line Segment Classifiers Rsari A. Medina Rdriguez and Rnald Fumi Hashimt University f Sa Paul Institute
More informationSAMPLING DYNAMICAL SYSTEMS
SAMPLING DYNAMICAL SYSTEMS Melvin J. Hinich Applied Research Labratries The University f Texas at Austin Austin, TX 78713-8029, USA (512) 835-3278 (Vice) 835-3259 (Fax) hinich@mail.la.utexas.edu ABSTRACT
More informationNeural Networks with Wavelet Based Denoising Layers for Time Series Prediction
Neural Netwrks with Wavelet Based Denising Layers fr Time Series Predictin UROS LOTRIC 1 AND ANDREJ DOBNIKAR University f Lublana, Faculty f Cmputer and Infrmatin Science, Slvenia, e-mail: {urs.ltric,
More informationFIELD QUALITY IN ACCELERATOR MAGNETS
FIELD QUALITY IN ACCELERATOR MAGNETS S. Russenschuck CERN, 1211 Geneva 23, Switzerland Abstract The field quality in the supercnducting magnets is expressed in terms f the cefficients f the Furier series
More informationPhysical Layer: Outline
18-: Intrductin t Telecmmunicatin Netwrks Lectures : Physical Layer Peter Steenkiste Spring 01 www.cs.cmu.edu/~prs/nets-ece Physical Layer: Outline Digital Representatin f Infrmatin Characterizatin f Cmmunicatin
More information1 Course Notes in Introductory Physics Jeffrey Seguritan
Intrductin & Kinematics I Intrductin Quickie Cncepts Units SI is standard system f units used t measure physical quantities. Base units that we use: meter (m) is standard unit f length kilgram (kg) is
More informationFabrication Thermal Test. Methodology for a Safe Cask Thermal Performance
ENSA (Grup SEPI) Fabricatin Thermal Test. Methdlgy fr a Safe Cask Thermal Perfrmance IAEA Internatinal Cnference n the Management f Spent Fuel frm Nuclear Pwer Reactrs An Integrated Apprach t the Back-End
More informationMath 105: Review for Exam I - Solutions
1. Let f(x) = 3 + x + 5. Math 105: Review fr Exam I - Slutins (a) What is the natural dmain f f? [ 5, ), which means all reals greater than r equal t 5 (b) What is the range f f? [3, ), which means all
More informationRay tracing equations in transversely isotropic media Cosmin Macesanu and Faruq Akbar, Seimax Technologies, Inc.
Ray tracing equatins in transversely istrpic media Csmin Macesanu and Faruq Akbar, Seimax Technlgies, Inc. SUMMARY We discuss a simple, cmpact apprach t deriving ray tracing equatins in transversely istrpic
More informationSurface and Contact Stress
Surface and Cntact Stress The cncept f the frce is fundamental t mechanics and many imprtant prblems can be cast in terms f frces nly, fr example the prblems cnsidered in Chapter. Hwever, mre sphisticated
More informationInternal vs. external validity. External validity. This section is based on Stock and Watson s Chapter 9.
Sectin 7 Mdel Assessment This sectin is based n Stck and Watsn s Chapter 9. Internal vs. external validity Internal validity refers t whether the analysis is valid fr the ppulatin and sample being studied.
More informationDistributions, spatial statistics and a Bayesian perspective
Distributins, spatial statistics and a Bayesian perspective Dug Nychka Natinal Center fr Atmspheric Research Distributins and densities Cnditinal distributins and Bayes Thm Bivariate nrmal Spatial statistics
More informationCurrent/voltage-mode third order quadrature oscillator employing two multiple outputs CCIIs and grounded capacitors
Indian Jurnal f Pure & Applied Physics Vl. 49 July 20 pp. 494-498 Current/vltage-mde third rder quadrature scillatr emplying tw multiple utputs CCIIs and grunded capacitrs Jiun-Wei Hrng Department f Electrnic
More informationCHAPTER 6 -- ENERGY. Approach #2: Using the component of mg along the line of d:
Slutins--Ch. 6 (Energy) CHAPTER 6 -- ENERGY 6.) The f.b.d. shwn t the right has been prvided t identify all the frces acting n the bdy as it mves up the incline. a.) T determine the wrk dne by gravity
More informationPHYS 314 HOMEWORK #3
PHYS 34 HOMEWORK #3 Due : 8 Feb. 07. A unifrm chain f mass M, lenth L and density λ (measured in k/m) hans s that its bttm link is just tuchin a scale. The chain is drpped frm rest nt the scale. What des
More informationCompetition and Invasion in a Microcosmic Setting
University f Tennessee, Knxville Trace: Tennessee Research and Creative Exchange University f Tennessee Hnrs Thesis Prjects University f Tennessee Hnrs Prgram 5-2004 Cmpetitin and Invasin in a Micrcsmic
More informationMethods for Determination of Mean Speckle Size in Simulated Speckle Pattern
0.478/msr-04-004 MEASUREMENT SCENCE REVEW, Vlume 4, N. 3, 04 Methds fr Determinatin f Mean Speckle Size in Simulated Speckle Pattern. Hamarvá, P. Šmíd, P. Hrváth, M. Hrabvský nstitute f Physics f the Academy
More informationSequential Allocation with Minimal Switching
In Cmputing Science and Statistics 28 (1996), pp. 567 572 Sequential Allcatin with Minimal Switching Quentin F. Stut 1 Janis Hardwick 1 EECS Dept., University f Michigan Statistics Dept., Purdue University
More informationFebruary 28, 2013 COMMENTS ON DIFFUSION, DIFFUSIVITY AND DERIVATION OF HYPERBOLIC EQUATIONS DESCRIBING THE DIFFUSION PHENOMENA
February 28, 2013 COMMENTS ON DIFFUSION, DIFFUSIVITY AND DERIVATION OF HYPERBOLIC EQUATIONS DESCRIBING THE DIFFUSION PHENOMENA Mental Experiment regarding 1D randm walk Cnsider a cntainer f gas in thermal
More informationWRITING THE REPORT. Organizing the report. Title Page. Table of Contents
WRITING THE REPORT Organizing the reprt Mst reprts shuld be rganized in the fllwing manner. Smetime there is a valid reasn t include extra chapters in within the bdy f the reprt. 1. Title page 2. Executive
More informationTrigonometric Ratios Unit 5 Tentative TEST date
1 U n i t 5 11U Date: Name: Trignmetric Ratis Unit 5 Tentative TEST date Big idea/learning Gals In this unit yu will extend yur knwledge f SOH CAH TOA t wrk with btuse and reflex angles. This extensin
More information