Training Algorithms for Recurrent Neural Networks

Size: px
Start display at page:

Download "Training Algorithms for Recurrent Neural Networks"

Transcription

1 raining Algrithms fr Recurrent Neural Netwrks SUWARIN PAAAVORAKUN UYN NGOC PIEN Cmputer Science Infrmatin anagement Prgram Asian Institute f echnlgy P.O. Bx 4, Klng Luang, Pathumthani AILAND Abstract: -Recurrent neural netwrks (RNNs), in which activity patterns pass thrugh the netwrk mre than nce befre they generate an utput pattern, can learn extremely cmplex tempral sequences. In this paper, three imprtant architectures f RNNs were described, alng with five existing training algrithms ne prpsed. An empirical study was made t evaluate the perfrmance f the frecasting mdels based n these netwrks the algrithms cnsidered, using the daily clsing stck prices f five prminent cmpanies listed n the Securities Exchange f hail. Frm the simulated results, ne may cnclude that gd frecasting mdels can be based n RNNs, the prpsed algrithm can perfrm very satisfactrily in terms f bth the frecast accuracy the cmputatin time. Key-Wrds: - Recurrent Neural Netwrks, Fully Recurrent Neural Netwrk, Lcally Recurrent Neural Netwrks, Efficiency Index, Stck Price Frecasting. 1 Intrductin Feedfrward neural netwrks are universal functin apprximatrs gd at detecting nnlinearies, but suffer frm lng training time a very high number f alternatives as far as architectures parameters g. hey are als prne t verfitting data. Recurrent neural netwrks (RNNs), in which the input layer s activity patterns pass thrugh the netwrk mre than nce befre generating a new utput pattern, can learn extremely cmplex tempral patterns. Several researchers have cnfirmed the superirity f RNNs ver feedfrward netwrks when perfrming nnlinear time series predictin [1, 2] fr an applicatin f RNNs t recgnitin f stck price patterns. RNNs require substantially mre cnnectins mre memry in simulatin than stard backprpagatin netwrks. RNNs can yield gd results because f the repetitin f similar patterns present in stck price time series. One f the main reasns fr being interested in RNNs is that there are new prpsed algrithms which allw them t learn hw t interact with an envirnment in an apprpriate way. he architecture f a predictr underpins its capacity t represent the dynamic prperties f the input signal hence its ability t frecast sme future value. rever, the stck market data nrmally react rapidly t the interventin f human being. his study fcuses n the selected training algrithms fr ppular RNN structures fr the predictin f the prices f the stcks f imprtant cmpanies listed n the Stck Exchange f hail. 2 Recurrent Neural Netwrks A frecasting mdel based n an RNN invlves the cnstructin f tw separate cmpnents : ne r mre recurrent layers (t prvide the tempral cntext, referred t as shrt-term memry) a predictr (usually the feedback part f the netwrk). his study cnsiders three variatins f the RNNs cnsisting the fully RNN tw lcal feedback netwrks, namely the Narendra-Parthasarathey Franscni-Gri-Sda netwrks. 2.1 Fully Recurrent Neural Netwrks (FRNNs) Shwn in Fig.1 is a typical FRNN where all the utputs f the existing neurns are used fr feedback its neurns are fully cnnected. he simplest memry element is the unit time delay, which has the transfer functin, (Z)= Z.

2 3 raining Algrithms he bjective f a training algrithm is t update the weights cnnecting t every nde f the utput units in rder t reach the minimum errr f the system. Fig. 1 Fully recurrent neural netwrk 2.2 Narendra-Parthasarathy (N-P) Netwrk his is a partial cnnectin with feedback links frm each nde f the utput layer t all hidden ndes [3]. In this netwrk, utput values are fed back t the netwrk as inputs act as shrt-term memry. Fig. 2 A Narendra-Parthasarathy recurrent netwrk 2.3 Franscni-Gri-Sda (FGS) Netwrk his is a lcally recurrent netwrk, where a multilayer perceptin is augmented with lcal feed back arund each hidden nde [4]. A ptential advantage f this netwrk is that it can adapt its internal representatin. Fig. 3 A Frascni-Gri-Sda lcally recurrent netwrk 3.1 Ntatin In a FRNN, m(n)-p-r tplgy, let y(k) r y k dente the n-tuple f utputs f the units in the netwrk at time k, let u(k) r u k dente the m-tuple f external input signals t the netwrk at time k. Each neurn in bth the hidden utput layers has a feedback lp frm the neurn itself the recurrent links frm ther neurns. distinguish the cmpnents f y representing unit utput frm thse representing external input values, let O dente the set f indicies i such that y i is the utput f a unit in the netwrk I dente the set f indicies i fr which y i is an external input. y i (k) = u i (k), i I = 1, 2,.., m y i (k) i O = 1, 2,.., n Let w ij dente the weight n the cnnectin t the i th unit frm either the j th unit, if j O r the j th input unit if j I. Let n = p + r w ij be the weight f link frm nde j t nde i, where, i = 1,.., r, r+1,, n = r+p; j = 1,., r, r+1,, n = r+p, n+1,, n + m. - N-P Netwrk: the algrithms fr N&P netwrk are the same as fr FRNN, but when updating the weights, we set w ij = 0, i = 1, 2,., n j = 1, 2,.,p. - FGS Netwrk: the algrithms fr FGS netwrk can be btained frm thse fr the FRNN, but when updating the weight, we set w ij = 0, i = 1, 2,., n j = 1, 2,.,m. 3.2 Basic Algrithms Y-N Algrithm: his algrithm, btained by Yamamt Nikifruk [5], incrprates an Errr Back Prpagatin (EBP) methd fr btaining a fictitius target signal f utput f the hidden ndes. he weight parameters are btained by an Expnentially Weighted Least Squares (EWLS) methd. A-P Algrithm: Atiya Parls [6] btained this algrithm by apprximating the errr gradient. his is an nline algrithm which assumes a small change in the netwrk state finds the directin f weight change by apprximatin. When minimizing the errr, the errr f hidden nde is equal zer, the errr f utput nde is the difference

3 between desired mdel utput. he determinatin f the change in the weights will be accrding t the change in netwrk state. he basic idea f the A-P algrithm is t interchange the rles f the netwrk states y(k) the weight matrix W. he states are cnsidered as the cntrl variables, the change in weights is determined accrding t the change in y(k). First dified Y-N Algrithm, An nline training algrithm was btained by cmbining three techniques, namely Errr Back Prpagatin (EBP), Recursive Least Squares (RLS), Errr Self Recurrent (ESR). his algrithm, dented YNC t recgnize the wrk f Chairatanatrai [7], updates weights bth in hidden ndes utput ndes in the same prcedure while the Y-N algrithm updates the weights f hidden ndes utput ndes separately. rever, the errrs calculated frm the utput unit are fed back fr determining weight updates f utput unit ndes. his part is called ESR. Secnd dified Algrithm he algrithm, dented YNS [8], is btained by cmbining the methds f weight updating f the A- P algrithm the Y-N algrithm techniques t enhance the netwrk perfrmance. Firstly weight cnnectins f the hidden ndes are adapted befre weight cnnectins f utput ndes in rder t generate net signal fr each hidden nde. Secndly, the netwrk uses these signals t adjust weight cnnectins f utput ndes. Frward Prpagatin Algrithm Als called real time recurrent learning (RRL) [9], this algrithm where the weights are updated in an n-line fashin. he gradients f the netwrk states with regard t the weights at time instant k are btained in terms f thse at time instant k. Once these are evaluated, the errr gradients can be btained in a straightfrward way. Prpsed Algrithm he algrithm is btained by cmbining the methds f weight update f the A-P algrithm (t find the directin f weight change by apprximatin), the Y-N algrithm (t estimate fictitius target signals f hidden ndes in rder t update hidden weight separately frm utput weights), then by adding the errr self-recurrent (ESR) netwrk t speed up the cnvergence. he prpsed algrithm, dented YN-, can be summarized as fllws: 1. Assume that data sets (y(1), d(1),..., y( ), d( )) is given. 2. Initialize the value f t = 1, let dente sme mathematical ntatins in rder t find weight changes as fllws γ () = -D ( )e () + We ( ) (1) B() = γ (t)y (t ) (2) t = 1 Because f ill-cnditining prblems, Eq.3 must be added with the uter prduct matrix by a small matrix ε I, where ε is a small psitive cnstant. V() = εi + = y(t)y (t) t Updated weights f hidden ndes are expressed by W = ηb 0 ()V () (3) (4) W (new) = W (ld) + W (5) hen, all new hidden utputs are calculated based n the updated hidden weights. y (t) = W y (t) = (t)y(t ) (y (t), y (t), u(t) ) (6) Finally, updated weights f utput ndes are expressed by W = ηb ()V () (7) W (new) = W (ld) + W (8) 3. Calculate a vectr f errr self-recurrent netwrks t be used as the errr signals fr utput hidden ndes. se(t) = e(t) + µ e(t) where se(t) = [se (t), se h (t),..,se (n) (t)], n-tuples are a number f utput units, µ is a cnstant f the time delayed errrs called ne-step previus errr, x (t) s (t) e(t) = x h (t) s h (t) 4. ime step at t = t+1, calculate the mdel utput. y(t) = f(wy(t )) (9) hen, cmpute the errr signal e (t) by ESR methd then using these signal in rder t generate as in Eq. 1 γ(t) 5. Find weighted changes f hidden ndes, W (t)

4 W (t) = W (t ) + γ (t)y (t )V (t η V ) - B (t )V (t )y(t ) 1 + y (t )V (t )y(t ) [ V (t )y(t )] [ V (t )y(t )][ V (t )y(t )] (10) (t) = V (t ) - 1+ y (t )V (t )y(t ) (11) B (t) = B (t ) + γ (t)y (t ) (12) 6. All new hidden utput signals are calculated based n the updated hidden weights. y (t) W (t)y(t ) = y (t) = (y (t), y (t), u(t) ) (13) 7. Find weighted changes f utput ndes, W (t) W (t) = W (t ) + γ (t)y (t ) η V (t) = V B V (t ) y (t ) V (t )y (t ) (t ) - B (t )V (t )y (t ) [ V (t )y (t )] [ V (t )y (t )][ V (t )y (t )] 1+ y (t ) V (t) = B (t ) + γ (t)y (t ) (t )y (t ) (14) (15) (16) 8. Repeat steps 3, 4, 5, 6 7 until the end f given data. erminating f training is terminated when the system errr has reached an acceptable criterin. Nte that set W W t zer at the beginning f next iteratin. - Prperty Cnstructin/Building furnishing materials sectr: Siam City Cement Public Cmpany Limited (SCCC). Daily data n the stck prices vlumes in the market frm 01 July 1999 t 30 June 2002 were used fr training phase 01 July June 2003 fr testing phase. he stck market is clsed n weekends, s the cnsecutive data series are frm ndays t Fridays. he frecasting equatin[10] with lead time f ne day can be expressed as: k P V i P(t+1) = g(p(t), P(t), P(t-2), V(t), V(t), V(t-2)) + h(p i (k), P p (k), P (k)) = time step f the netwrk = stck price in baht = stck vlume in unit = hidden nde i ( i = 1, 2, p). = utput nde Befre being presented t the RNN, the data are transfrmed by a linear affine transfrmatin t the range [0.05, 0.95] using the equatins: [ 0.9(y - y )/(y - y )] 0.05 y = (17) t t min max min + [(y max - y min )(y t )/0.9] y min y t = + (18) where y t y t dente the transfrmed riginal data, ymax ymin are the maximum minimum f the riginal data in the calibratin phase. A stpping rule fr indicating the cnvergence f the algrithms cnsidered. It is based n the relatinship between the sum f squared errrs f tw cnsecutive iteratins, it is as fllws: 4 Applicatins Discussins SE(t + 1) - SE(t) SE(t) ε (19) 4.1 Data Emplyed Stck prices data in hai arket data were btained frm he Securities Exchange f hail (SE). Five imprtant cmpanies were selected: - Real Estate Sectr: L use Public Cmpany Limited (L) - Banking Sectr :Bangkk Bank Public Cmpany Limited (BBL) Kasikrn Bank Public Cmpany Limited (KBANK), - Cmmunicatin Sectr: Shin Crpratin Public Cmpany Limited (SIN), where SE(t+1) is the Sum f Squared Errrs at iteratin (t+1), ε is a small number, set equal t in the present study. measure the accuracy f frecasting values, the efficiency index (EI) prpsed by Nash Sutcliffee [11] can be used. SE EI = 1 (20) S 2 2 S = (yi - y), SE = = (yi - ŷ i ), i = 1 i 1

5 y = 1 i = 1 y i where S = tal variatin, SE = Sum f squared errrs, y i = Actual utput, i.e. bserved value at time i, y = ean value f the actual utput, yˆ i = del utput, i.e. frecast value at time i, = Number f data pints A value f EI clse t 1.0 indicates an excellent perfrmance in terms f the predictin accuracy. 4.2 Experimental Cases Cnditins he training algrithms n the three RNN mdels are evaluated in terms f the cmputatin time the frecast accuracy expressed by EI. Fr these data sets, the 6(3)-2 tplgy is used. his tplgy has 6 external inputs (3 inputs fr stck price, 3 inputs fr the vlume), 3 feedback lines frm hidden ndes ne utput nde. All feedback cnnectin weights were set equal 1.0. he sigmid slpe f sigmid functin was set equal t 1. he same initial weights in the range f [0, 1] were used Fr the Y-N algrithm, the frgetting factr (β) was set at 0.99, Fr the A-P YNS algrithms, parameter η was set at 0.009, parameter ε at 13.0 fr L, KBANK SIN at 8.5 fr BBL SCCC, Fr the YNC algrithm, the bias f ne-step previus errr (µ ) was set at 1.0, Fr the RRL algrithm, the learning rate (η) at 0.003, Fr the YN- algrithm, µ was set at 0.3, η at ε at 13.0 fr L, KBANK SIN at 8.5 fr BBL SCCC. 4.3 Results Discussins he values f the efficiency index btained in the training phase are cllected in able 1 thse f the cmputatin time in able 2. able 1. Cmputed values f efficiency index fr stck prices in hai market Stck FRNN Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Stck N-P Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Stck FGS Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Frm able 1, bth the N-P FGS netwrks prduce the values f EI less than thse prduced by the FRNN. his may be due t less infrmatin fr capturing the variability (less weight cnnectins). - - Amng six algrithms, the perfrmance f the RRL algrithm, in terms f frecasting accuracy, is the lwest, while the YN- algrithm appears t be the best. he Y-N,YNC, YNS YN- algrithms result in quite high values f EI with the FRNN mdel, but slightly lwer with the N-P FGS netwrks. he prpsed algrithm, YN- appears t perfrm very well fr all netwrk architectures. - In view f the cmputatin time (able 2), the prpsed algrithm als perfrms very well. It is the least time cnsuming amng the algrithms cnsidered. It shuld be nted that the values used fr the parameters f the different algrithms d have great influence n the time cnsumptin. he different algrithms have different sensitive parameters which must be tuned first befre training the netwrks. In this study, the apprpriate parameter values f each algrithm were chsen fixed, then the netwrks were trained many times with different initial weights in the range frm 0 t 1. Afterwards, the same initial weights were fixed fr training all the netwrks.

6 able 2. Cmputatinal time (in secnds) Stck FRNN Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Stck N-P Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Stck FGS Cmpany Y-N A-P YNC YNS RRL YN- L BBL KBANK SIN SCCC Fr each algrithm, bth N-P FGS netwrks require lwer cmputatin time than the FRNN des. his is due t the fact that these tw netwrks have simpler architectures with limited feedback lps. As a trade-ff, these netwrks d nt prduce high values fr the efficiency index as mentined earlier. 5 Cnclusins A brief descriptin f three recurrent netwrk architecture was given in this study. It is fllwed by a brief descriptin f five existing training algrithms ne prpsed in this study. By an empirical study using the stck data frm five imprtant cmpanies listed n the Securities Exchange f hail, the fllwing cnclusins may be drawn: - Frecasting shrt-term stck prices based n the histrical daily clsing prices is pssible with the use f RNN mdels. - he prpsed algrithm appears t perfrm very well in terms f bth the frecast accuracy the cmputatin time. As such, it can be used fr training these RNN mdels. References: [1] K. Kamij, &. anigawa, Stck Price Pattem Recgnitin : A Recurrent Neural Netwrk Apprach. Prceedings f the Internatinal Jint Cnference n Neural Netwrks. 1990, [2] C.L. Giles, S.Lawrence & A.C. si, Rule Inference fr Financial Predictin using Recurrent Neural Netwrks, IEEE Cnference n Cmputatinal Intelligence fr Financial Engineering, IEEE Press, 1997, pp.253. [3] K.S.Narendra, & K. Parthasarathy, Identificatin Cntrl f Dynamical Systems using Neural Netwrks, IEEE ransactins n Neural Netwrks, Vl 1, N. 1, 1990, pp [4] P. Franscni,. Gri, & G. Sda, Lcal Feedback ultilayered Netwrks Neural Cmputatin, Vl. 4, N.1, 1992, pp [5] Y. Yamamt, & P.N. Nikifruk, A Learning Algrithm fr Recurrent Neural Netwrks its Applicatin t Nnlinear Identificatin, Prceedings f the 1999 IEEE Internatinal Sympsium n Cmputer Aided Cntrl System Design, awai, USA., August, pp [6] A.F. Atiya, & A.G. Parls, New Result n Recurrent Netwrk raining : Unifying the Algrithms Accelerating Cnvergence, IEEE ransactins n Neural Netwrks, Vl.11, N.3, pp [7] A. Chairatanatrai, A Cmparisn f Selected raining Algrithms fr Recurrent Neural Netwrks, aster hesis., Asian Institute f echnlgy, Pathumthani, hail, [8] S. Pattamavrakun &.N. Phien, A Futher Study n Recurrent Neural Neural Netwrks fr Frecasting, Prc.Cnf. he 8 th Wrld ulti- Cnference n Systemics Cybernetics Infrmics, USA, ( appear) [9] R.J. William, & P.Zipser, A Learning Algrithm fr Cntinually Running Fully Recurrent Neural Netwrks Neural Cmputatin, Vl. 1, N.12, 1989, pp [10] N.. Danh,.N.Phien & A.D. Gupta, Neural Netwrk dels fr River Flw Frecasting, Water SA, Vl.25 N.1, 33-39, [11] J.E. Nash, & J. V. Sutcliffee, River Flw Frecasting thrugh Cnceptual dels, Jurnal f ydrlgy, Vl. 10, 1970, pp

the results to larger systems due to prop'erties of the projection algorithm. First, the number of hidden nodes must

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 information

A Scalable Recurrent Neural Network Framework for Model-free

A 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 information

Enhancing Performance of MLP/RBF Neural Classifiers via an Multivariate Data Distribution Scheme

Enhancing 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 information

1996 Engineering Systems Design and Analysis Conference, Montpellier, France, July 1-4, 1996, Vol. 7, pp

1996 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 information

Learning to Control an Unstable System with Forward Modeling

Learning to Control an Unstable System with Forward Modeling 324 Jrdan and Jacbs Learning t Cntrl an Unstable System with Frward Mdeling Michael I. Jrdan Brain and Cgnitive Sciences MIT Cambridge, MA 02139 Rbert A. Jacbs Cmputer and Infrmatin Sciences University

More information

On Huntsberger Type Shrinkage Estimator for the Mean of Normal Distribution ABSTRACT INTRODUCTION

On 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 information

Pattern Recognition 2014 Support Vector Machines

Pattern 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 information

Design and Simulation of Dc-Dc Voltage Converters Using Matlab/Simulink

Design 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 information

Dataflow Analysis and Abstract Interpretation

Dataflow Analysis and Abstract Interpretation Dataflw Analysis and Abstract Interpretatin Cmputer Science and Artificial Intelligence Labratry MIT Nvember 9, 2015 Recap Last time we develped frm first principles an algrithm t derive invariants. Key

More information

Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeoff

Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeoff Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeff Reading: Chapter 2 STATS 202: Data mining and analysis September 27, 2017 1 / 20 Supervised vs. unsupervised learning In unsupervised

More information

NUROP CONGRESS PAPER CHINESE PINYIN TO CHINESE CHARACTER CONVERSION

NUROP 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 information

Revision: August 19, E Main Suite D Pullman, WA (509) Voice and Fax

Revision: 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 information

Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeoff

Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeoff Lecture 2: Supervised vs. unsupervised learning, bias-variance tradeff Reading: Chapter 2 STATS 202: Data mining and analysis September 27, 2017 1 / 20 Supervised vs. unsupervised learning In unsupervised

More information

Technical Bulletin. Generation Interconnection Procedures. Revisions to Cluster 4, Phase 1 Study Methodology

Technical 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 information

Artificial Neural Networks MLP, Backpropagation

Artificial 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 information

Chapter 3: Cluster Analysis

Chapter 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 information

Least Squares Optimal Filtering with Multirate Observations

Least 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 information

Computational modeling techniques

Computational 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 information

BASD HIGH SCHOOL FORMAL LAB REPORT

BASD HIGH SCHOOL FORMAL LAB REPORT BASD HIGH SCHOOL FORMAL LAB REPORT *WARNING: After an explanatin f what t include in each sectin, there is an example f hw the sectin might lk using a sample experiment Keep in mind, the sample lab used

More information

EXPERIMENTAL STUDY ON DISCHARGE COEFFICIENT OF OUTFLOW OPENING FOR PREDICTING CROSS-VENTILATION FLOW RATE

EXPERIMENTAL STUDY ON DISCHARGE COEFFICIENT OF OUTFLOW OPENING FOR PREDICTING CROSS-VENTILATION FLOW RATE EXPERIMENTAL STUD ON DISCHARGE COEFFICIENT OF OUTFLOW OPENING FOR PREDICTING CROSS-VENTILATION FLOW RATE Tmnbu Gt, Masaaki Ohba, Takashi Kurabuchi 2, Tmyuki End 3, shihik Akamine 4, and Tshihir Nnaka 2

More information

APPLICATION OF THE BRATSETH SCHEME FOR HIGH LATITUDE INTERMITTENT DATA ASSIMILATION USING THE PSU/NCAR MM5 MESOSCALE MODEL

APPLICATION OF THE BRATSETH SCHEME FOR HIGH LATITUDE INTERMITTENT DATA ASSIMILATION USING THE PSU/NCAR MM5 MESOSCALE MODEL JP2.11 APPLICATION OF THE BRATSETH SCHEME FOR HIGH LATITUDE INTERMITTENT DATA ASSIMILATION USING THE PSU/NCAR MM5 MESOSCALE MODEL Xingang Fan * and Jeffrey S. Tilley University f Alaska Fairbanks, Fairbanks,

More information

Optimization Programming Problems For Control And Management Of Bacterial Disease With Two Stage Growth/Spread Among Plants

Optimization 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 information

CHAPTER 24: INFERENCE IN REGRESSION. Chapter 24: Make inferences about the population from which the sample data came.

CHAPTER 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 information

Fall 2013 Physics 172 Recitation 3 Momentum and Springs

Fall 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 information

Aerodynamic Separability in Tip Speed Ratio and Separability in Wind Speed- a Comparison

Aerodynamic Separability in Tip Speed Ratio and Separability in Wind Speed- a Comparison Jurnal f Physics: Cnference Series OPEN ACCESS Aerdynamic Separability in Tip Speed Rati and Separability in Wind Speed- a Cmparisn T cite this article: M L Gala Sants et al 14 J. Phys.: Cnf. Ser. 555

More information

Building to Transformations on Coordinate Axis Grade 5: Geometry Graph points on the coordinate plane to solve real-world and mathematical problems.

Building 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 information

k-nearest Neighbor How to choose k Average of k points more reliable when: Large k: noise in attributes +o o noise in class labels

k-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 information

Dead-beat controller design

Dead-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 information

Methods for Determination of Mean Speckle Size in Simulated Speckle Pattern

Methods 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 information

Analysis on the Stability of Reservoir Soil Slope Based on Fuzzy Artificial Neural Network

Analysis 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 information

Multiple Source Multiple. using Network Coding

Multiple 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 information

ROUNDING ERRORS IN BEAM-TRACKING CALCULATIONS

ROUNDING 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 information

Admin. MDP Search Trees. Optimal Quantities. Reinforcement Learning

Admin. 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 information

initially 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

initially 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 information

Eric Klein and Ning Sa

Eric Klein and Ning Sa Week 12. Statistical Appraches t Netwrks: p1 and p* Wasserman and Faust Chapter 15: Statistical Analysis f Single Relatinal Netwrks There are fur tasks in psitinal analysis: 1) Define Equivalence 2) Measure

More information

Bootstrap 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) >

Bootstrap 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 information

SAMPLING DYNAMICAL SYSTEMS

SAMPLING 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 information

Determining the Accuracy of Modal Parameter Estimation Methods

Determining 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 information

ChE 471: LECTURE 4 Fall 2003

ChE 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 information

Module 4: General Formulation of Electric Circuit Theory

Module 4: General Formulation of Electric Circuit Theory Mdule 4: General Frmulatin f Electric Circuit Thery 4. General Frmulatin f Electric Circuit Thery All electrmagnetic phenmena are described at a fundamental level by Maxwell's equatins and the assciated

More information

Perfrmance f Sensitizing Rules n Shewhart Cntrl Charts with Autcrrelated Data Key Wrds: Autregressive, Mving Average, Runs Tests, Shewhart Cntrl Chart

Perfrmance f Sensitizing Rules n Shewhart Cntrl Charts with Autcrrelated Data Key Wrds: Autregressive, Mving Average, Runs Tests, Shewhart Cntrl Chart Perfrmance f Sensitizing Rules n Shewhart Cntrl Charts with Autcrrelated Data Sandy D. Balkin Dennis K. J. Lin y Pennsylvania State University, University Park, PA 16802 Sandy Balkin is a graduate student

More information

Slide04 (supplemental) Haykin Chapter 4 (both 2nd and 3rd ed): Multi-Layer Perceptrons

Slide04 (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 information

Supplementary Course Notes Adding and Subtracting AC Voltages and Currents

Supplementary Course Notes Adding and Subtracting AC Voltages and Currents Supplementary Curse Ntes Adding and Subtracting AC Vltages and Currents As mentined previusly, when cmbining DC vltages r currents, we nly need t knw the plarity (vltage) and directin (current). In the

More information

MODULE FOUR. This module addresses functions. SC Academic Elementary Algebra Standards:

MODULE 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 information

Numerical Simulation of the Thermal Resposne Test Within the Comsol Multiphysics Environment

Numerical Simulation of the Thermal Resposne Test Within the Comsol Multiphysics Environment Presented at the COMSOL Cnference 2008 Hannver University f Parma Department f Industrial Engineering Numerical Simulatin f the Thermal Respsne Test Within the Cmsl Multiphysics Envirnment Authr : C. Crradi,

More information

Distributions, spatial statistics and a Bayesian perspective

Distributions, 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 information

CHAPTER 4 DIAGNOSTICS FOR INFLUENTIAL OBSERVATIONS

CHAPTER 4 DIAGNOSTICS FOR INFLUENTIAL OBSERVATIONS CHAPTER 4 DIAGNOSTICS FOR INFLUENTIAL OBSERVATIONS 1 Influential bservatins are bservatins whse presence in the data can have a distrting effect n the parameter estimates and pssibly the entire analysis,

More information

CAUSAL INFERENCE. Technical Track Session I. Phillippe Leite. The World Bank

CAUSAL 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 information

Cambridge Assessment International Education Cambridge Ordinary Level. Published

Cambridge 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 )

[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 information

SPH3U1 Lesson 06 Kinematics

SPH3U1 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 information

Linearization of the Output of a Wheatstone Bridge for Single Active Sensor. Madhu Mohan N., Geetha T., Sankaran P. and Jagadeesh Kumar V.

Linearization 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 information

What is Statistical Learning?

What is Statistical Learning? What is Statistical Learning? Sales 5 10 15 20 25 Sales 5 10 15 20 25 Sales 5 10 15 20 25 0 50 100 200 300 TV 0 10 20 30 40 50 Radi 0 20 40 60 80 100 Newspaper Shwn are Sales vs TV, Radi and Newspaper,

More information

ENSC Discrete Time Systems. Project Outline. Semester

ENSC 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 information

Lead/Lag Compensator Frequency Domain Properties and Design Methods

Lead/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 information

COMP 551 Applied Machine Learning Lecture 9: Support Vector Machines (cont d)

COMP 551 Applied Machine Learning Lecture 9: Support Vector Machines (cont d) COMP 551 Applied Machine Learning Lecture 9: Supprt Vectr Machines (cnt d) Instructr: Herke van Hf (herke.vanhf@mail.mcgill.ca) Slides mstly by: Class web page: www.cs.mcgill.ca/~hvanh2/cmp551 Unless therwise

More information

8 th Grade Math: Pre-Algebra

8 th Grade Math: Pre-Algebra Hardin Cunty Middle Schl (2013-2014) 1 8 th Grade Math: Pre-Algebra Curse Descriptin The purpse f this curse is t enhance student understanding, participatin, and real-life applicatin f middle-schl mathematics

More information

Modeling the Nonlinear Rheological Behavior of Materials with a Hyper-Exponential Type Function

Modeling the Nonlinear Rheological Behavior of Materials with a Hyper-Exponential Type Function www.ccsenet.rg/mer Mechanical Engineering Research Vl. 1, N. 1; December 011 Mdeling the Nnlinear Rhelgical Behavir f Materials with a Hyper-Expnential Type Functin Marc Delphin Mnsia Département de Physique,

More information

5 th grade Common Core Standards

5 th grade Common Core Standards 5 th grade Cmmn Cre Standards In Grade 5, instructinal time shuld fcus n three critical areas: (1) develping fluency with additin and subtractin f fractins, and develping understanding f the multiplicatin

More information

Current/voltage-mode third order quadrature oscillator employing two multiple outputs CCIIs and grounded capacitors

Current/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 information

Kinetic Model Completeness

Kinetic 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 information

CHAPTER 3 INEQUALITIES. Copyright -The Institute of Chartered Accountants of India

CHAPTER 3 INEQUALITIES. Copyright -The Institute of Chartered Accountants of India CHAPTER 3 INEQUALITIES Cpyright -The Institute f Chartered Accuntants f India INEQUALITIES LEARNING OBJECTIVES One f the widely used decisin making prblems, nwadays, is t decide n the ptimal mix f scarce

More information

Floating Point Method for Solving Transportation. Problems with Additional Constraints

Floating Point Method for Solving Transportation. Problems with Additional Constraints Internatinal Mathematical Frum, Vl. 6, 20, n. 40, 983-992 Flating Pint Methd fr Slving Transprtatin Prblems with Additinal Cnstraints P. Pandian and D. Anuradha Department f Mathematics, Schl f Advanced

More information

Math Foundations 20 Work Plan

Math Foundations 20 Work Plan Math Fundatins 20 Wrk Plan Units / Tpics 20.8 Demnstrate understanding f systems f linear inequalities in tw variables. Time Frame December 1-3 weeks 6-10 Majr Learning Indicatrs Identify situatins relevant

More information

Differentiation Applications 1: Related Rates

Differentiation Applications 1: Related Rates Differentiatin Applicatins 1: Related Rates 151 Differentiatin Applicatins 1: Related Rates Mdel 1: Sliding Ladder 10 ladder y 10 ladder 10 ladder A 10 ft ladder is leaning against a wall when the bttm

More information

Physical Layer: Outline

Physical 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 information

Early detection of mining truck failure by modelling its operation with neural networks classification algorithms

Early detection of mining truck failure by modelling its operation with neural networks classification algorithms RU, Rand GOLOSINSKI, T.S. Early detectin f mining truck failure by mdelling its peratin with neural netwrks classificatin algrithms. Applicatin f Cmputers and Operatins Research ill the Minerals Industries,

More information

Section I5: Feedback in Operational Amplifiers

Section 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 information

Physics 2010 Motion with Constant Acceleration Experiment 1

Physics 2010 Motion with Constant Acceleration Experiment 1 . Physics 00 Mtin with Cnstant Acceleratin Experiment In this lab, we will study the mtin f a glider as it accelerates dwnhill n a tilted air track. The glider is supprted ver the air track by a cushin

More information

Comparison of two variable parameter Muskingum methods

Comparison of two variable parameter Muskingum methods Extreme Hydrlgical Events: Precipitatin, Flds and Drughts (Prceedings f the Ykhama Sympsium, July 1993). IAHS Publ. n. 213, 1993. 129 Cmparisn f tw variable parameter Muskingum methds M. PERUMAL Department

More information

MINIMIZATION 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 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 information

COMP9444 Neural Networks and Deep Learning 3. Backpropagation

COMP9444 Neural Networks and Deep Learning 3. Backpropagation COMP9444 Neural Netwrks and Deep Learning 3. Backprpagatin Tetbk, Sectins 4.3, 5.2, 6.5.2 COMP9444 17s2 Backprpagatin 1 Outline Supervised Learning Ockham s Razr (5.2) Multi-Layer Netwrks Gradient Descent

More information

CS 477/677 Analysis of Algorithms Fall 2007 Dr. George Bebis Course Project Due Date: 11/29/2007

CS 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 information

ENG2410 Digital Design Sequential Circuits: Part A

ENG2410 Digital Design Sequential Circuits: Part A ENG2410 Digital Design Sequential Circuits: Part A Fall 2017 S. Areibi Schl f Engineering University f Guelph Week #6 Tpics Sequential Circuit Definitins Latches Flip-Flps Delays in Sequential Circuits

More information

Churn Prediction using Dynamic RFM-Augmented node2vec

Churn Prediction using Dynamic RFM-Augmented node2vec Churn Predictin using Dynamic RFM-Augmented nde2vec Sandra Mitrvić, Jchen de Weerdt, Bart Baesens & Wilfried Lemahieu Department f Decisin Sciences and Infrmatin Management, KU Leuven 18 September 2017,

More information

Performance Bounds for Detect and Avoid Signal Sensing

Performance Bounds for Detect and Avoid Signal Sensing Perfrmance unds fr Detect and Avid Signal Sensing Sam Reisenfeld Real-ime Infrmatin etwrks, University f echnlgy, Sydney, radway, SW 007, Australia samr@uts.edu.au Abstract Detect and Avid (DAA) is a Cgnitive

More information

Computational modeling techniques

Computational 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 information

Chapter 3 Kinematics in Two Dimensions; Vectors

Chapter 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 information

SUPPLEMENTARY MATERIAL GaGa: a simple and flexible hierarchical model for microarray data analysis

SUPPLEMENTARY MATERIAL GaGa: a simple and flexible hierarchical model for microarray data analysis SUPPLEMENTARY MATERIAL GaGa: a simple and flexible hierarchical mdel fr micrarray data analysis David Rssell Department f Bistatistics M.D. Andersn Cancer Center, Hustn, TX 77030, USA rsselldavid@gmail.cm

More information

Resampling Methods. Chapter 5. Chapter 5 1 / 52

Resampling Methods. Chapter 5. Chapter 5 1 / 52 Resampling Methds Chapter 5 Chapter 5 1 / 52 1 51 Validatin set apprach 2 52 Crss validatin 3 53 Btstrap Chapter 5 2 / 52 Abut Resampling An imprtant statistical tl Pretending the data as ppulatin and

More information

Neural Networks with Wavelet Based Denoising Layers for Time Series Prediction

Neural 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 information

Tree Structured Classifier

Tree Structured Classifier Tree Structured Classifier Reference: Classificatin and Regressin Trees by L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stne, Chapman & Hall, 98. A Medical Eample (CART): Predict high risk patients

More information

A Correlation of. to the. South Carolina Academic Standards for Mathematics Precalculus

A Correlation of. to the. South Carolina Academic Standards for Mathematics Precalculus A Crrelatin f Suth Carlina Academic Standards fr Mathematics Precalculus INTRODUCTION This dcument demnstrates hw Precalculus (Blitzer), 4 th Editin 010, meets the indicatrs f the. Crrelatin page references

More information

Soliton-Effect Optical Pulse Compression in Bulk Media with χ (3) Nonlinearity. 1 Introduction

Soliton-Effect Optical Pulse Compression in Bulk Media with χ (3) Nonlinearity. 1 Introduction Nnlinear Analysis: Mdelling and Cntrl, Vilnius, IMI,, N 5 Lithuanian Assciatin f Nnlinear Analysts, Slitn-Effect Optical Pulse Cmpressin in Bulk Media with χ (3) Nnlinearity Received: 9.7. Accepted: 11.1.

More information

Sample questions to support inquiry with students:

Sample questions to support inquiry with students: Area f Learning: Mathematics Calculus 12 Big Ideas Elabratins The cncept f a limit is fundatinal t calculus. cncept f a limit: Differentiatin and integratin are defined using limits. Sample questins t

More information

We can see from the graph above that the intersection is, i.e., [ ).

We 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 information

The blessing of dimensionality for kernel methods

The blessing of dimensionality for kernel methods fr kernel methds Building classifiers in high dimensinal space Pierre Dupnt Pierre.Dupnt@ucluvain.be Classifiers define decisin surfaces in sme feature space where the data is either initially represented

More information

Department of Electrical Engineering, University of Waterloo. Introduction

Department of Electrical Engineering, University of Waterloo. Introduction Sectin 4: Sequential Circuits Majr Tpics Types f sequential circuits Flip-flps Analysis f clcked sequential circuits Mre and Mealy machines Design f clcked sequential circuits State transitin design methd

More information

A Matrix Representation of Panel Data

A 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 information

DYNAMIC MODELLING OF N-CARDAN TRANSMISSIONS WITH SHAFTS IN SPATIAL CONFIGURATION. Part II. THE ALGORITHM OF DYNAMIC MODELLING

DYNAMIC MODELLING OF N-CARDAN TRANSMISSIONS WITH SHAFTS IN SPATIAL CONFIGURATION. Part II. THE ALGORITHM OF DYNAMIC MODELLING Fascicle f Management and Technlgical Engineering, Vlume VI (XVI), 7 DYNAMIC MODELLING OF N-CARDAN TRANSMISSIONS WITH SHAFTS IN SPATIAL CONFIGURATION. Part II. THE ALGORITHM OF DYNAMIC MODELLING Cdrua

More information

Supplementary Course Notes Adding and Subtracting AC Voltages and Currents

Supplementary Course Notes Adding and Subtracting AC Voltages and Currents Supplementary Curse Ntes Adding and Subtracting AC Vltages and Currents As mentined previusly, when cmbining DC vltages r currents, we nly need t knw the plarity (vltage) and directin (current). In the

More information

3.4 Shrinkage Methods Prostate Cancer Data Example (Continued) Ridge Regression

3.4 Shrinkage Methods Prostate Cancer Data Example (Continued) Ridge Regression 3.3.4 Prstate Cancer Data Example (Cntinued) 3.4 Shrinkage Methds 61 Table 3.3 shws the cefficients frm a number f different selectin and shrinkage methds. They are best-subset selectin using an all-subsets

More information

On Boussinesq's problem

On Boussinesq's problem Internatinal Jurnal f Engineering Science 39 (2001) 317±322 www.elsevier.cm/lcate/ijengsci On Bussinesq's prblem A.P.S. Selvadurai * Department f Civil Engineering and Applied Mechanics, McGill University,

More information

A mathematical model for complete stress-strain curve prediction of permeable concrete

A mathematical model for complete stress-strain curve prediction of permeable concrete A mathematical mdel fr cmplete stress-strain curve predictin f permeable cncrete M. K. Hussin Y. Zhuge F. Bullen W. P. Lkuge Faculty f Engineering and Surveying, University f Suthern Queensland, Twmba,

More information

Application of APW Pseudopotential Form Factor in the Calculation of Liquid Metal Resistivities.

Application of APW Pseudopotential Form Factor in the Calculation of Liquid Metal Resistivities. Internatinal Jurnal f Pure and Applied Physics. ISSN 097-1776 Vlume 8, Number (01), pp. 11-117 Research India Publicatins http://www.ripublicatin.cm/pap.htm Applicatin f APW Pseudptential Frm Factr in

More information

Relationships Between Frequency, Capacitance, Inductance and Reactance.

Relationships 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 information

o o IMPORTANT REMINDERS Reports will be graded largely on their ability to clearly communicate results and important conclusions.

o o IMPORTANT REMINDERS Reports will be graded largely on their ability to clearly communicate results and important conclusions. BASD High Schl Frmal Lab Reprt GENERAL INFORMATION 12 pt Times New Rman fnt Duble-spaced, if required by yur teacher 1 inch margins n all sides (tp, bttm, left, and right) Always write in third persn (avid

More information

ENG2410 Digital Design Sequential Circuits: Part B

ENG2410 Digital Design Sequential Circuits: Part B ENG24 Digital Design Sequential Circuits: Part B Fall 27 S. Areibi Schl f Engineering University f Guelph Analysis f Sequential Circuits Earlier we learned hw t analyze cmbinatinal circuits We will extend

More information

Resampling Methods. Cross-validation, Bootstrapping. Marek Petrik 2/21/2017

Resampling 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 information