Advanced Street Lighting Control through Neural Network Ensembling

Size: px
Start display at page:

Download "Advanced Street Lighting Control through Neural Network Ensembling"

Transcription

1 Advanced Street Lghtng Control through Neural Network Ensemblng Stefano Pzzut, Fabo Morett, Mauro Annunzato ENEA (Energy New technology and Envronment Agency) Rome, Italy Stefano Panzer Unverstà Roma Tre DIA (Department of Informatcs and Automaton) Rome, Italy Abstract In ths work, we propose an nnovatve street lghtng energy management system n order to reduce energy consumpton. The man goal s to provde energy on demand such that energy, n ths case lght, s provded only when needed. In order to acheve ths purpose t s crtcal to have a relable demand model, whch n the case of street lghtng turns out to be a traffc flow rate forecastng model. Several methods has been compared n order to fnd out one hour predcton model. In our case studes, Artfcal Neural Networks performed best results. Moreover, several control strateges have been tested and the one whch gave the best energy savngs s the adaptve one we carred out. Expermentaton has been carred out on two dfferent case studes. In partcular we focused our expermentaton on publc street of a small and a medum szed ctes. Our studes show that wth the proposed approach t s possble to save up to 50% of energy compared to no regulaton systems Keywords-Lghtng Effcency; Energy Management Systems; Adaptve Control; Neural-Network models I. INTRODUCTION Snce the frst nternatonal recommendatons for the lghtng of roads [1], power consumpton and envronmental aspects have become more and more mportant and at the same tme, the mproved performance of lumnares and lamps, and especally the ntroducton of electronc control gears, has made t possble to ntroduce adaptve lghtng for motorzed roads and pedestrans areas. A structured model has been developed for the selecton of the approprate lghtng classes [2] (M, C, or P), based on the lumnance concept, takng nto account the dfferent parameters relevant for the gven vsual tasks. Applyng for example tme dependent varables lke traffc volume or weather condtons, the model offers the possblty to use adaptve lghtng systems wth remarkable energy consumpton savngs and therefore hgh fnancal benefts for those muncpaltes [3] where street lghtng s a hgh percentage of the electrcal bll. Today, lghtng control approaches ranges from smple on/off to regulaton systems. On/off systems nclude tmers, twlght and astronomcal clocks. The frst one s a statc system whch turns on and off street lghts always accordng to fxed tmes. The second one has lght-senstve photocells to turn them on at dusk and off at dawn [5]. The thrd ones are GPS based street lght controllers whch operate the on/off of the street lght accordng to the locaton features (longtude, lattude, sunrse, sunset tmes). Regulatons systems are based on dmmable LED or hgh pressure sodum vapor lghts [4] and allow to schedule lghts on or off and set dmmng levels of ndvdual or groups of lghts. All these systems have one common feature: they do not care about the real on-lne demand and ths s a source of hgh neffcency. Thus, n order to overcome the man lack of the current regulaton systems, t has recently started the new Intellgent Street Lghtng (ISL) approach whch looks very promsng [5]-[6]. Therefore, here we propose an ISL approach (Smart Adaptve Control) based on the concept of energy on demand, whose goal s to dynamcally set the lght ntensty as functon of the foreseen demand, namely the traffc flow rate 1 hour forecast. Thus, n such context the demand model has a crtcal role and ts accuracy strongly affects the performance of the regulaton system. In the last decade, one of the most wdely used method n order to solve modelng problems s that of Artfcal Neural Networks (ANN) [7]-[8]. In partcular, traffc flow forecastng ssue has been tackled through ANN snce the nnetes [9]-[19]. As example, among the most recent work [19] focuses on traffc flow forecastng approach based on Partcle Swarm Optmzaton (PSO) wth Wavelet Network Model (WNM). Pamula [16] revews neural networks applcatons n urban traffc management systems and presents a method of traffc flow predcton based on neural networks. Bucur et al. [17] proposes the use of a selfadaptve fuzzy neural network for traffc predcton suggestng an archtecture whch tracks probablty dstrbuton drfts due to weather condtons, season, or other factors. All the mentoned applcatons have one feature n common: they use one sngle global model n order to perform the predcton. Our approach s to use not only one model but an ensemble of models. In Secton II an overvew of modelng 76

2 methods used s gven, n partcular statstcal and Artfcal Neural Network based models and ther combnaton trough ensemblng. In Secton III we show results obtaned on two case studes and then n Secton IV we dscuss future works. used to model f a are the hyperbolc tangent, the sgmod and the lnear functon. II. MODELING METHODS In ths secton, we shortly descrbe the modelng and control technques we compared n the expermentaton. A. Statstcal Modelng One the smplest and most wdely used model s to buld an average weekly dstrbuton of the traffc flow rate sampled hourly. Thus, from the data we compute for each day the average traffc flow rate hour by hour n such a way that we get an average dstrbuton made of 24X7=168 ponts. B. Artfcal Neural Networks Artfcal Neural Networks (ANN) are computatonal models whch try to smulate some propertes of bologcal neural networks n order to solve complex modelng problems of non-lnear systems. An ANN s an nterconnected group of artfcal neurons (called also nodes) that uses a mathematcal or computatonal model for nformaton processng based on a connectonstc approach to computaton. In more practcal terms ANN are non-lnear data modelng or decson makng tools whch can be used to model complex relatonshps between nputs and outputs or to fnd patterns n data. ANN are referred also as blackbox or data-drven models and they are manly used when analytcal or transparent models cannot be appled. Buldng such models needs several stages as nput analyss and tranng through algorthms whch mnmze the error between the real values to be modeled and the ANN output. ANN demonstrated ther effectveness n modelng many real-world applcatons. Once we model an ANN model, we must take nto account three basc components. Frst, the synapses of the bologcal neuron are modeled as weghts. Let us remember that the synapse of the bologcal neuron s the one whch nterconnects the neural network and gves the strength of the connecton. For an artfcal neuron, the weght s a number, and represents the synapse. A negatve weght reflects an nhbtory connecton, whle postve values desgnate exctatory connectons. The followng components of the model represent the actual actvty of the neuron cell. All nputs are summed altogether and modfed by the weghts. Ths actvty s referred as a lnear combnaton. Fnally, an actvaton functon controls the ampltude of the output. Mathematcally, ths process s descrbed n Fgure 1. From ths model the actvty of the neuron can be shown to be: y=f a (w x - ) (1) where s a threshold called BIAS (Basc Input Actvaton System) whch dentfes the senstvty of the neuron to respond to the external nputs. The most common functon Fgure 1. Artfcal Neuron Model Therefore each unt performs a relatvely smple job: receve nput from neghbors or external sources and use ths to compute an output sgnal whch s propagated to other unts. Apart from ths processng, a second task s the adjustment of the weghts. The system s nherently parallel n the sense that many unts can carry out ther computatons at the same tme. Wthn neural systems t s useful to dstngush three types of unts: nput unts whch receve data from outsde the neural network, output unts whch send data out of the neural network, and hdden unts whose nput and output sgnals reman wthn the network. The way unts are connected defnes the network topology or archtecture. In the past years, many of them have been studed and the most wdely used and s the feed-forward one. In ths network structure, neurons are grouped nto layers. There are at least two layers, the nput and the output, whch gather the correspondng nput and output varables. Ths basc structure s also known as perceptron [20]. Fgure 2. Feed-forward neural network topology Moreover, n order to let the model cope wth non-lnear problems, t s possble to add one or more ntermedate layers, known as hdden layers. These models are also known as mult-layer perceptrons (MLP) [21]. The flow of data from nput to output unts s strctly n one drecton (forward). The data processng can extend over 77

3 multple (layers of) unts, but no feedback connectons are present, that s, connectons extendng from outputs of unts to nputs of unts n the same layer or prevous layers. C. Ensemblng The term ensemble descrbes a group of learnng machnes that work together on the same task, n the case of ANN they are traned on some data, run together and ther outputs are combned as a sngle one. The goal s obtan better predctve performance than could be obtaned from any of the consttuent models. A. Modelng Street Street Street The basc dea s to set the power level of the followng hour as functon of the ANN ensemble forecast. P t+1 = f ( t+1 ) (2) where P t+1 s the power level normalzed n [0,1] to be set for the next hour, t+1 s the traffc flow rate neural forecast whch s t+1 = anne( t, t-1,, t-n ) (3) where anne s the ANN ensemble result, t- s the measured traffc flow rate at tme t-. For Street lghtng applcatons functon f n (2) can be shaped n dfferent ways, among these we appled a lnear profle although nternatonal standards [2] suggest a nonlnear one that we wll apply n future work. Fgure 3. Ensemblng In the last years several ensemblng methods have been carred out [22],[23],[24]. The frst one, also known as Basc Ensemble Method (BEM), s the smplest way to combne M neural networks as an arthmetc mean of ther outputs y. Ths method can mprove the global performance [25],[26] although t does not takes nto account that some models can be more accurate than others. Ths method has the advantage to be very easy to apply. A drect BEM extenson s the Generalsed Ensemble Method (GEM) [25],[26] n whch the outputs of the sngle models are combned n a weghted average where the weghts have to be properly set, sometmes after an expensve tunng process. Other methods are Bootstrap AGGregatING (BAGGING) [27] and Adaboost [28],[29]. III. EXPERIMENTATIONS In ths secton, we test and compare the methods presented n the prevous sectons. We used two test cases: one has concerned Tern and the second regards S.Govann n Persceto. In the frst one, we focused on three dfferent urban streets of Tern (Table I). The data set s made of 3 months (13 weeks) of measurement correspondng to 2184 hourly samples. The data set has been parttoned nto tranng/testng and valdaton made respectvely of 10 and 3 weeks each. TABLE I. STREET FEATURES Maxmum traffc flow rate If k < 0.25 then f = 0.5 If k > 0.5 then f = 1 Else f = 2 k (4) where k s the predcted traffc flow rate at tme k (3) normalzed n [0,1]. The ANN are feed-forward MLP wth 10 hdden neurons and one output (the one hour flow forecast) wth sgmod as actvaton functon for all the neurons. The number of nputs s the same of the dynamcs wndow length (3) and t has been chosen wth a prelmnary analyss by calculatng the valdaton predcton error, after the ensemblng stage, for dfferent number of hourly samples (Table II). Snce we obtaned the mnmum predcton error wth a tme wndow of eght hours, we chose the same number of nputs for the ANN, representng the length of tme hstory wndow. So each nput contans the traffc flow of one of eght hours of the tme wndow. Tranng has been performed through the Back-Propagaton algorthm wth adaptve learnng rate and momentum stoppng after 108 teratons and a save best strategy to avod overfttng. The reported results are averaged over 10 dfferent runs (wth standard devaton n brackets) and the ensemble s therefore made by the same 10 models. The reported errors are measured as: e = x-y /(M-m) (5) Where x s the real value to be predcted, y s the output model, M s the real maxmum value and m s the mnmum. 78

4 TABLE II. WINDOW HISTORY LENGTH (HOURS) SELECTION Number of Street 1 Street 2 Street 3 Samples % 6.88% 5.81% 5 3.9% 5.07% 3.99% % 3.43% 3.02% % 4.12% 3.74% At last, Table III shows the comparson of the models consdered n ths work n terms of predcton accuracy over the valdaton set. Fgure 4 shows a graphcal comparson. We compared real hourly traffc flow rate wth predcton provded from statstcal and neural network ensemble models. Ensemble models outperforms statstcal n all cases. TABLE III. MODEL COMPARISON Statstc ANN ANN Ensemblng Street % 3.74% 3.29% (±0.10%) Street % 3.48% 3.02% (±0.09%) Street % 4.00% 3.43% (±0.10%) Average 6.20% 3.74% (±0.10%) 3.25% From ths analyss t s clear that n general the ensemblng approach outperforms the statstcal approach provdng a remarkable mprovement n predcton accuracy. Such level of precson s very mportant when dealng wth applcatons lke traffc and lghtng control where the hgher the model accuracy s the more effectve the control system s. From ths graph t s clear that the ANN ensemble model performs much better than the statstcal model because, when out of normal condtons the ANN ensemblng takes nto account the real traffc dynamcs (3). B. Control In ths secton, we compare the results of the Statc Control (StaC) and the Smart Adaptve Control (SmAC) ntroduced n Secton III. In the expermentaton we calculated the energy savng of the two methods wth respect to the no-regulaton strategy, namely when lghts are always on at 100% of ther power for the whole nght. The lght on demand control assumes dmmable lghts (SAP or LED), on the streets we carred out ths study there were no such lamps and data about the real consumptons were not avalable, therefore the expermentaton has been carred out off-lne by calculatng the potental energy consumptons n the followng way. It has been assumed the maxmum hourly nomnal power consumpton to be one, then the followng quanttes have been calculated : C 100 = x 1, x 1 0, 1 (6) Where x 1 s the hourly power level for the th sample accordng to the no control strategy (nght power level always at 100%) and therefore C 100 s ts overall consumpton. C StaC = x 2, x 2 0, 0.5, 1 (7) Where x 2 s the hourly power level for the th sample accordng to the statc control (StaC) strategy (Fg. 4) and therefore C StaC s ts overall consumpton. C SmAC = x 3, x 3 [0, 1] (8) Where x 3 s the hourly power level for the th sample (4) accordng to the smart adaptve control (SmAC) strategy (Fg. 5) and therefore C SmAC s ts overall consumpton. These quanttes have been calculated over three months of actual traffc flow rates obtaned by on street col sensors. Thus, we computed the consumpton savng of the StaC and SmAC strateges wth respect to the no control approach n the followng way : S StaC = 1 - C StaC / C 100 (9) S SmAC = 1 - C SmAC / C 100 (10) In Table IV, we report these values for the three consdered streets. Fgure 4. Models comparson 79

5 TABLE IV. CONTROL STRATEGY COMPARISON: ENERGY SAVING StaC SmAC Street 1 25% 44.5% Street 2 25% 47% Street 3 25% 37.5% Average 25% 43% Actual hour: next traffc flow predcton based on prevous hour traffc flow Prevous week: forecast based on the same hour and the same week day of the prevous week Statstc model: averaged hourly profle Neural Ensemblng: neural network ensemblng based model Results show that t s possble to save on average 43% of energy, meanng that lamps wll work at 57% of ther nomnal power havng as nferor lmt 50% (4) and Fg. 5 n order to avod perods durng normal operaton wth almost no lght due to lght output drop. From these results t s clear that the SmAC approach provdes a remarkable mprovement n terms of energy savng (43% on average) n partcular on streets wth medum-low traffc flow rate. Moreover, n Fgure 5 t s shown an example of how the two strateges work, where on the Y axs we report the normalzed traffc flow rate values and the normalzed hourly power consumptons of the dfferent strateges. From Fgure 5 t s possble to see that the SmAC strategy s capable to follow the real demand (traffc flow rate) achevng the energy on demand concept. In partcular, t s nterestng to pont out that SmAC mproves not only energy effcency (orange dotted area) but also safety (yellow dashed area) because t provdes lght when actually needed. TABEL V. COMPARISON OF FORECASTING MODELS ERRORS Actual Hour. Prevous week Statstc model Neural Ensemblng Error 8.77% 7% 5.53% 4.39% Results show that also n ths case Neural Ensemble outperforms other methods. Then we compared three dfferent control descrbed above: constant, statc and adaptve. As shown n Fg. 6 constant control does not take n account the traffc flow, statc control does not overcome the varablty of traffc flow meanwhle adaptve control provde energy to lghtng spot proportonal to traffc flow. Fgure 6. Comparson of control strateges TABLE VI. CONTROL STRATEGY COMPARISON (ENERGY SAVING) Statc Control Adaptve Control -12% -34% Energy savng show n Table VI s a theoretcal evaluaton respect of constant control. Fgure 5. Control strategy comparson Tests performed on S.Govann n Persceto are based on a dataset of 123 days, sampled hourly, for a total of 2952 hours. Once agan we compared dfferent forecastng system: IV. CONCLUSIONS AND FUTURE WORKS In ths work, we proposed a new approach for adaptve street lghtng control based on the energy on demand dea. In order to acheve ths goal t s crtcal to have a relable demand model, whch n the case of street lghtng turns out to be a traffc flow rate forecastng model. Thus, we showed a modelng approach based on Artfcal Neural Networks Ensemblng n order to provde a one hour forecast of urban traffc flow rates. Expermentaton has been carred out on three dfferent classes of real streets and 80

6 results showed that the proposed approach clearly outperforms the statstcal methods (6% predcton error) achevng a 3% predcton error. The reason for that s that the neural ensemblng model s capable to provde more relable estmatons when out of standard condtons because t consders the real traffc dynamcs. Moreover, the proposed adaptve control strategy has been tested and compared to a tradtonal regulaton system on the same streets. Results showed that the adaptve control provdes, on average, energy savngs almost doubled (43% vs 25%). Future work wll frstly focus on dmmng profles accordng to nternatonal standards, then further modelng mprovements (usng more sophstcated ensemblng methods as well as tryng to develop hybrd models) wll be nvestgated and lastly, the economc mpact of the proposed methodology wll be carred out. Moreover further forecastng model can be taken n account n order to valdate the qualty of results obtaned. REFERENCES [1] CIE 115, Recommendatons for the Lghtng of Roads for Motor and Pedestran Traffc, [2] CIE 115, Lghtng of Roads for Motor and Pedestran Traffc, [3] Transport, Envronmental and communty servces commttee. Reducng Energy Consumpton - Street Lghtng. Report No TEC/87/08. The Hghland Councl, [4] M.S.Rea, J.D. Bullough, Y. Akash, "Several vews of metal halde and hgh pressure sodum lghtng for outdoor applcatons". Lghtng Research and Technology. p. 41(4): pp , [5] Intellgent Street Lghtng Saves Up to 80% On Energy, ScenceDaly, July [6] New System of Intellgent Management of Street Lghtng Enables 80% Savngs n Energy, ScenceDaly, November [7] M.A. Arbb. The Handbook of Bran Theory and Neural Networks, The MIT Press, Cambrdge (MA), [8] S. Haykn. Neural Networks, A comprehensve foundaton" (2nd edton), Neural Networks, Prentce Hall, New Jersey, [9] D. Canca, J. Larrañeta, S. Lozano, L.Oneva, Traffc ntensty forecast n urban networks usng a multlayer perceptron. Jont Internatonal Meetng, Barcelona [10] M.S. Dougherty and M.Cobbett, Short term nter-urban traffc forecasts usng neural networks. Proc. Second DRIVE-II Workshop on Short-Term Traffc Forecastng, Delft, The Netherlands, pp , [11] S.Ishak, P. Kotha, C.Alecsandru, Optmzaton of Dynamc Neural Network Performance for Short-Term Traffc Predcton. In Transportaton Research Record, 1836, pp [12] D. Park, L.R. Rlett, G. Han Spectral Bass Neural Networks for Real-Tme Travel Tme Forecastng, Journal of Transportaton Engneerng 125(6), 1999, pp [13] C. Taylor and D.Meldrum, Freeway traffc data predcton usng neural networks. Proc. 6th VNIS Conference, Seattle, WA, pp , [14] J.W.C. Van Lnt, S.P. Hoogendoorn, H.J. Van Zuylen, Freeway Travel tme Predcton wth State-Space Neural Networks, Natonal Academes Press, Transportaton Research Record 1811, pp ,, Washngton D.C., [15] W.Zheng, L. Der-Horng, Q.Sh, Short-Term Freeway Traffc Flow Pre-dcton: Bayesan Combned Neural Network Approach, Journal of Transportaton Engneerng, 2006, pp [16] T. Pamula, Road traffc parameters predcton n urban traffc management systems usng neural networks, Transport Problems, Volume 6 Issue 3, 2011, pp [17] L. Bucur, A.Florea, B.S.Petrescu, An adaptve fuzzy neural network for traffc predcton, Control & Automaton (MED), 18th Medterranean Conference, Marrakech, pp , [18] B.G. Çetner, M.Sar, Q.Borat, A neural network based traffc flow predcton model, Mathematcal and Computatonal Applcatons, Vol. 15, No. 2, 2010, pp [19] S.Jawanjal, P. Bajaj, A desgn approach to traffc flow forecastng wth soft computng tools, Emergng Trends n Engneerng and Technology (ICETET), 3rd Internatonal Conference, pp , Goa, [20] F. Rosenblatt, The perceptron: a percevng and recognzng automaton. Report , Cornell Aeronautcal Laboratory, [21] F. Rosenblatt, Prncples of neurodynamcs: perceptrons and the theory of bran mechansms. Spartan Books, Washngton D.C.,1961. [22] A. Krogh, J. Vedelsby, Neural network ensembles, cross valdaton and actve learnng, Advances n Neural Informaton Processng Systems, 7, pp , MIT Press, [23] Y. Lu, X. Yao, Ensemble learnng va negatve correlaton, Neural Networks, 12(10), 1999, pp [24] L. Breman, Combnng Predctors, Combnng Artfcal Neural Nets - Ensemble and Modular Mult-net Systems, Sprnger, Berln, pp , [25] M.P. Perrone and L.N. Cooper, When networks dsagree: ensemble methods for hybrd neural networks. Neural Networks for Speech and Image Processng, Chapman-Hall, [26] C.M. Bshop, Neural networks for pattern recognton. Oxford Unversty Press, pp , [27] R. Kohav and E. Bauer, An emprcal comparson of votng classfcaton algorthms: Baggng, boostng and varants, Machne Learnng, 36, 1999, pp [28] H. Drucker, Improvng regressors usng boostng technques, Machne Learnng, Proceedngs of the Fourteenth Internatonal Conference, pp , [29] R. Avnmelech and N. Intrator, Boostng regresson estmators. Neural Computaton, 11, 1999, pp

Neural Networks & Learning

Neural Networks & Learning Neural Netorks & Learnng. Introducton The basc prelmnares nvolved n the Artfcal Neural Netorks (ANN) are descrbed n secton. An Artfcal Neural Netorks (ANN) s an nformaton-processng paradgm that nspred

More information

Generalized Linear Methods

Generalized Linear Methods Generalzed Lnear Methods 1 Introducton In the Ensemble Methods the general dea s that usng a combnaton of several weak learner one could make a better learner. More formally, assume that we have a set

More information

Uncertainty in measurements of power and energy on power networks

Uncertainty in measurements of power and energy on power networks Uncertanty n measurements of power and energy on power networks E. Manov, N. Kolev Department of Measurement and Instrumentaton, Techncal Unversty Sofa, bul. Klment Ohrdsk No8, bl., 000 Sofa, Bulgara Tel./fax:

More information

Boostrapaggregating (Bagging)

Boostrapaggregating (Bagging) Boostrapaggregatng (Baggng) An ensemble meta-algorthm desgned to mprove the stablty and accuracy of machne learnng algorthms Can be used n both regresson and classfcaton Reduces varance and helps to avod

More information

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results.

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results. Neural Networks : Dervaton compled by Alvn Wan from Professor Jtendra Malk s lecture Ths type of computaton s called deep learnng and s the most popular method for many problems, such as computer vson

More information

Admin NEURAL NETWORKS. Perceptron learning algorithm. Our Nervous System 10/25/16. Assignment 7. Class 11/22. Schedule for the rest of the semester

Admin NEURAL NETWORKS. Perceptron learning algorithm. Our Nervous System 10/25/16. Assignment 7. Class 11/22. Schedule for the rest of the semester 0/25/6 Admn Assgnment 7 Class /22 Schedule for the rest of the semester NEURAL NETWORKS Davd Kauchak CS58 Fall 206 Perceptron learnng algorthm Our Nervous System repeat untl convergence (or for some #

More information

Supporting Information

Supporting Information Supportng Informaton The neural network f n Eq. 1 s gven by: f x l = ReLU W atom x l + b atom, 2 where ReLU s the element-wse rectfed lnear unt, 21.e., ReLUx = max0, x, W atom R d d s the weght matrx to

More information

Short Term Load Forecasting using an Artificial Neural Network

Short Term Load Forecasting using an Artificial Neural Network Short Term Load Forecastng usng an Artfcal Neural Network D. Kown 1, M. Km 1, C. Hong 1,, S. Cho 2 1 Department of Computer Scence, Sangmyung Unversty, Seoul, Korea 2 Department of Energy Grd, Sangmyung

More information

Appendix B: Resampling Algorithms

Appendix B: Resampling Algorithms 407 Appendx B: Resamplng Algorthms A common problem of all partcle flters s the degeneracy of weghts, whch conssts of the unbounded ncrease of the varance of the mportance weghts ω [ ] of the partcles

More information

EEE 241: Linear Systems

EEE 241: Linear Systems EEE : Lnear Systems Summary #: Backpropagaton BACKPROPAGATION The perceptron rule as well as the Wdrow Hoff learnng were desgned to tran sngle layer networks. They suffer from the same dsadvantage: they

More information

A LINEAR PROGRAM TO COMPARE MULTIPLE GROSS CREDIT LOSS FORECASTS. Dr. Derald E. Wentzien, Wesley College, (302) ,

A LINEAR PROGRAM TO COMPARE MULTIPLE GROSS CREDIT LOSS FORECASTS. Dr. Derald E. Wentzien, Wesley College, (302) , A LINEAR PROGRAM TO COMPARE MULTIPLE GROSS CREDIT LOSS FORECASTS Dr. Derald E. Wentzen, Wesley College, (302) 736-2574, wentzde@wesley.edu ABSTRACT A lnear programmng model s developed and used to compare

More information

Neural Networks. Perceptrons and Backpropagation. Silke Bussen-Heyen. 5th of Novemeber Universität Bremen Fachbereich 3. Neural Networks 1 / 17

Neural Networks. Perceptrons and Backpropagation. Silke Bussen-Heyen. 5th of Novemeber Universität Bremen Fachbereich 3. Neural Networks 1 / 17 Neural Networks Perceptrons and Backpropagaton Slke Bussen-Heyen Unverstät Bremen Fachberech 3 5th of Novemeber 2012 Neural Networks 1 / 17 Contents 1 Introducton 2 Unts 3 Network structure 4 Snglelayer

More information

Week 5: Neural Networks

Week 5: Neural Networks Week 5: Neural Networks Instructor: Sergey Levne Neural Networks Summary In the prevous lecture, we saw how we can construct neural networks by extendng logstc regresson. Neural networks consst of multple

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Lecture 23: Artificial neural networks

Lecture 23: Artificial neural networks Lecture 23: Artfcal neural networks Broad feld that has developed over the past 20 to 30 years Confluence of statstcal mechancs, appled math, bology and computers Orgnal motvaton: mathematcal modelng of

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

Multilayer Perceptron (MLP)

Multilayer Perceptron (MLP) Multlayer Perceptron (MLP) Seungjn Cho Department of Computer Scence and Engneerng Pohang Unversty of Scence and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjn@postech.ac.kr 1 / 20 Outlne

More information

Evaluation of classifiers MLPs

Evaluation of classifiers MLPs Lecture Evaluaton of classfers MLPs Mlos Hausrecht mlos@cs.ptt.edu 539 Sennott Square Evaluaton For any data set e use to test the model e can buld a confuson matrx: Counts of examples th: class label

More information

Solving Nonlinear Differential Equations by a Neural Network Method

Solving Nonlinear Differential Equations by a Neural Network Method Solvng Nonlnear Dfferental Equatons by a Neural Network Method Luce P. Aarts and Peter Van der Veer Delft Unversty of Technology, Faculty of Cvlengneerng and Geoscences, Secton of Cvlengneerng Informatcs,

More information

Chapter 9: Statistical Inference and the Relationship between Two Variables

Chapter 9: Statistical Inference and the Relationship between Two Variables Chapter 9: Statstcal Inference and the Relatonshp between Two Varables Key Words The Regresson Model The Sample Regresson Equaton The Pearson Correlaton Coeffcent Learnng Outcomes After studyng ths chapter,

More information

Computing Correlated Equilibria in Multi-Player Games

Computing Correlated Equilibria in Multi-Player Games Computng Correlated Equlbra n Mult-Player Games Chrstos H. Papadmtrou Presented by Zhanxang Huang December 7th, 2005 1 The Author Dr. Chrstos H. Papadmtrou CS professor at UC Berkley (taught at Harvard,

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

Model of Neurons. CS 416 Artificial Intelligence. Early History of Neural Nets. Cybernetics. McCulloch-Pitts Neurons. Hebbian Modification.

Model of Neurons. CS 416 Artificial Intelligence. Early History of Neural Nets. Cybernetics. McCulloch-Pitts Neurons. Hebbian Modification. Page 1 Model of Neurons CS 416 Artfcal Intellgence Lecture 18 Neural Nets Chapter 20 Multple nputs/dendrtes (~10,000!!!) Cell body/soma performs computaton Sngle output/axon Computaton s typcally modeled

More information

Multilayer Perceptrons and Backpropagation. Perceptrons. Recap: Perceptrons. Informatics 1 CG: Lecture 6. Mirella Lapata

Multilayer Perceptrons and Backpropagation. Perceptrons. Recap: Perceptrons. Informatics 1 CG: Lecture 6. Mirella Lapata Multlayer Perceptrons and Informatcs CG: Lecture 6 Mrella Lapata School of Informatcs Unversty of Ednburgh mlap@nf.ed.ac.uk Readng: Kevn Gurney s Introducton to Neural Networks, Chapters 5 6.5 January,

More information

Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application Peng-fei LIU1,Qun-tai SHEN1 and Jun ZHI2,*

Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application Peng-fei LIU1,Qun-tai SHEN1 and Jun ZHI2,* Advances n Computer Scence Research (ACRS), volume 54 Internatonal Conference on Computer Networks and Communcaton Technology (CNCT206) Usng Immune Genetc Algorthm to Optmze BP Neural Network and Its Applcaton

More information

Multigradient for Neural Networks for Equalizers 1

Multigradient for Neural Networks for Equalizers 1 Multgradent for Neural Netorks for Equalzers 1 Chulhee ee, Jnook Go and Heeyoung Km Department of Electrcal and Electronc Engneerng Yonse Unversty 134 Shnchon-Dong, Seodaemun-Ku, Seoul 1-749, Korea ABSTRACT

More information

Chapter 11: Simple Linear Regression and Correlation

Chapter 11: Simple Linear Regression and Correlation Chapter 11: Smple Lnear Regresson and Correlaton 11-1 Emprcal Models 11-2 Smple Lnear Regresson 11-3 Propertes of the Least Squares Estmators 11-4 Hypothess Test n Smple Lnear Regresson 11-4.1 Use of t-tests

More information

Comparison of Regression Lines

Comparison of Regression Lines STATGRAPHICS Rev. 9/13/2013 Comparson of Regresson Lnes Summary... 1 Data Input... 3 Analyss Summary... 4 Plot of Ftted Model... 6 Condtonal Sums of Squares... 6 Analyss Optons... 7 Forecasts... 8 Confdence

More information

BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS. M. Krishna Reddy, B. Naveen Kumar and Y. Ramu

BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS. M. Krishna Reddy, B. Naveen Kumar and Y. Ramu BOOTSTRAP METHOD FOR TESTING OF EQUALITY OF SEVERAL MEANS M. Krshna Reddy, B. Naveen Kumar and Y. Ramu Department of Statstcs, Osmana Unversty, Hyderabad -500 007, Inda. nanbyrozu@gmal.com, ramu0@gmal.com

More information

Statistics for Economics & Business

Statistics for Economics & Business Statstcs for Economcs & Busness Smple Lnear Regresson Learnng Objectves In ths chapter, you learn: How to use regresson analyss to predct the value of a dependent varable based on an ndependent varable

More information

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM Internatonal Conference on Ceramcs, Bkaner, Inda Internatonal Journal of Modern Physcs: Conference Seres Vol. 22 (2013) 757 761 World Scentfc Publshng Company DOI: 10.1142/S2010194513010982 FUZZY GOAL

More information

A New Algorithm for Training Multi-layered Morphological Networks

A New Algorithm for Training Multi-layered Morphological Networks A New Algorthm for Tranng Mult-layered Morphologcal Networs Rcardo Barrón, Humberto Sossa, and Benamín Cruz Centro de Investgacón en Computacón-IPN Av. Juan de Dos Bátz esquna con Mguel Othón de Mendzábal

More information

A Hybrid Variational Iteration Method for Blasius Equation

A Hybrid Variational Iteration Method for Blasius Equation Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 1932-9466 Vol. 10, Issue 1 (June 2015), pp. 223-229 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) A Hybrd Varatonal Iteraton Method

More information

Turbulence classification of load data by the frequency and severity of wind gusts. Oscar Moñux, DEWI GmbH Kevin Bleibler, DEWI GmbH

Turbulence classification of load data by the frequency and severity of wind gusts. Oscar Moñux, DEWI GmbH Kevin Bleibler, DEWI GmbH Turbulence classfcaton of load data by the frequency and severty of wnd gusts Introducton Oscar Moñux, DEWI GmbH Kevn Blebler, DEWI GmbH Durng the wnd turbne developng process, one of the most mportant

More information

Support Vector Machines. Vibhav Gogate The University of Texas at dallas

Support Vector Machines. Vibhav Gogate The University of Texas at dallas Support Vector Machnes Vbhav Gogate he Unversty of exas at dallas What We have Learned So Far? 1. Decson rees. Naïve Bayes 3. Lnear Regresson 4. Logstc Regresson 5. Perceptron 6. Neural networks 7. K-Nearest

More information

Annexes. EC.1. Cycle-base move illustration. EC.2. Problem Instances

Annexes. EC.1. Cycle-base move illustration. EC.2. Problem Instances ec Annexes Ths Annex frst llustrates a cycle-based move n the dynamc-block generaton tabu search. It then dsplays the characterstcs of the nstance sets, followed by detaled results of the parametercalbraton

More information

Chapter 3 Describing Data Using Numerical Measures

Chapter 3 Describing Data Using Numerical Measures Chapter 3 Student Lecture Notes 3-1 Chapter 3 Descrbng Data Usng Numercal Measures Fall 2006 Fundamentals of Busness Statstcs 1 Chapter Goals To establsh the usefulness of summary measures of data. The

More information

The Study of Teaching-learning-based Optimization Algorithm

The Study of Teaching-learning-based Optimization Algorithm Advanced Scence and Technology Letters Vol. (AST 06), pp.05- http://dx.do.org/0.57/astl.06. The Study of Teachng-learnng-based Optmzaton Algorthm u Sun, Yan fu, Lele Kong, Haolang Q,, Helongang Insttute

More information

Non-linear Canonical Correlation Analysis Using a RBF Network

Non-linear Canonical Correlation Analysis Using a RBF Network ESANN' proceedngs - European Smposum on Artfcal Neural Networks Bruges (Belgum), 4-6 Aprl, d-sde publ., ISBN -97--, pp. 57-5 Non-lnear Canoncal Correlaton Analss Usng a RBF Network Sukhbnder Kumar, Elane

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

III. Econometric Methodology Regression Analysis

III. Econometric Methodology Regression Analysis Page Econ07 Appled Econometrcs Topc : An Overvew of Regresson Analyss (Studenmund, Chapter ) I. The Nature and Scope of Econometrcs. Lot s of defntons of econometrcs. Nobel Prze Commttee Paul Samuelson,

More information

A New Evolutionary Computation Based Approach for Learning Bayesian Network

A New Evolutionary Computation Based Approach for Learning Bayesian Network Avalable onlne at www.scencedrect.com Proceda Engneerng 15 (2011) 4026 4030 Advanced n Control Engneerng and Informaton Scence A New Evolutonary Computaton Based Approach for Learnng Bayesan Network Yungang

More information

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing Advanced Scence and Technology Letters, pp.164-168 http://dx.do.org/10.14257/astl.2013 Pop-Clc Nose Detecton Usng Inter-Frame Correlaton for Improved Portable Audtory Sensng Dong Yun Lee, Kwang Myung Jeon,

More information

Wavelet chaotic neural networks and their application to continuous function optimization

Wavelet chaotic neural networks and their application to continuous function optimization Vol., No.3, 04-09 (009) do:0.436/ns.009.307 Natural Scence Wavelet chaotc neural networks and ther applcaton to contnuous functon optmzaton Ja-Ha Zhang, Yao-Qun Xu College of Electrcal and Automatc Engneerng,

More information

A Robust Method for Calculating the Correlation Coefficient

A Robust Method for Calculating the Correlation Coefficient A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal

More information

Regularized Discriminant Analysis for Face Recognition

Regularized Discriminant Analysis for Face Recognition 1 Regularzed Dscrmnant Analyss for Face Recognton Itz Pma, Mayer Aladem Department of Electrcal and Computer Engneerng, Ben-Guron Unversty of the Negev P.O.Box 653, Beer-Sheva, 845, Israel. Abstract Ths

More information

SOLVING CAPACITATED VEHICLE ROUTING PROBLEMS WITH TIME WINDOWS BY GOAL PROGRAMMING APPROACH

SOLVING CAPACITATED VEHICLE ROUTING PROBLEMS WITH TIME WINDOWS BY GOAL PROGRAMMING APPROACH Proceedngs of IICMA 2013 Research Topc, pp. xx-xx. SOLVIG CAPACITATED VEHICLE ROUTIG PROBLEMS WITH TIME WIDOWS BY GOAL PROGRAMMIG APPROACH ATMII DHORURI 1, EMIUGROHO RATA SARI 2, AD DWI LESTARI 3 1Department

More information

9 Derivation of Rate Equations from Single-Cell Conductance (Hodgkin-Huxley-like) Equations

9 Derivation of Rate Equations from Single-Cell Conductance (Hodgkin-Huxley-like) Equations Physcs 171/271 - Chapter 9R -Davd Klenfeld - Fall 2005 9 Dervaton of Rate Equatons from Sngle-Cell Conductance (Hodgkn-Huxley-lke) Equatons We consder a network of many neurons, each of whch obeys a set

More information

Lecture 4. Instructor: Haipeng Luo

Lecture 4. Instructor: Haipeng Luo Lecture 4 Instructor: Hapeng Luo In the followng lectures, we focus on the expert problem and study more adaptve algorthms. Although Hedge s proven to be worst-case optmal, one may wonder how well t would

More information

Ensemble Methods: Boosting

Ensemble Methods: Boosting Ensemble Methods: Boostng Ncholas Ruozz Unversty of Texas at Dallas Based on the sldes of Vbhav Gogate and Rob Schapre Last Tme Varance reducton va baggng Generate new tranng data sets by samplng wth replacement

More information

Feature Selection: Part 1

Feature Selection: Part 1 CSE 546: Machne Learnng Lecture 5 Feature Selecton: Part 1 Instructor: Sham Kakade 1 Regresson n the hgh dmensonal settng How do we learn when the number of features d s greater than the sample sze n?

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method

More information

Internet Engineering. Jacek Mazurkiewicz, PhD Softcomputing. Part 3: Recurrent Artificial Neural Networks Self-Organising Artificial Neural Networks

Internet Engineering. Jacek Mazurkiewicz, PhD Softcomputing. Part 3: Recurrent Artificial Neural Networks Self-Organising Artificial Neural Networks Internet Engneerng Jacek Mazurkewcz, PhD Softcomputng Part 3: Recurrent Artfcal Neural Networks Self-Organsng Artfcal Neural Networks Recurrent Artfcal Neural Networks Feedback sgnals between neurons Dynamc

More information

MAXIMUM A POSTERIORI TRANSDUCTION

MAXIMUM A POSTERIORI TRANSDUCTION MAXIMUM A POSTERIORI TRANSDUCTION LI-WEI WANG, JU-FU FENG School of Mathematcal Scences, Peng Unversty, Bejng, 0087, Chna Center for Informaton Scences, Peng Unversty, Bejng, 0087, Chna E-MIAL: {wanglw,

More information

1 Derivation of Rate Equations from Single-Cell Conductance (Hodgkin-Huxley-like) Equations

1 Derivation of Rate Equations from Single-Cell Conductance (Hodgkin-Huxley-like) Equations Physcs 171/271 -Davd Klenfeld - Fall 2005 (revsed Wnter 2011) 1 Dervaton of Rate Equatons from Sngle-Cell Conductance (Hodgkn-Huxley-lke) Equatons We consder a network of many neurons, each of whch obeys

More information

COMPARATIVE STUDY OF RAINFALL FORECASTING MODELS. Mohita Anand Sharma 1 and Jai Bhagwan Singh 2

COMPARATIVE STUDY OF RAINFALL FORECASTING MODELS. Mohita Anand Sharma 1 and Jai Bhagwan Singh 2 New York Scence Journal, 2011;4(7) http://www.scencepub.net/newyork COMPARATIVE STUDY OF RAINFALL FORECASTING MODELS Mohta Anand Sharma 1 and Ja Bhagwan Sngh 2 1. Research Scholar, 2. Senor Professor Statstcs.

More information

Uncertainty as the Overlap of Alternate Conditional Distributions

Uncertainty as the Overlap of Alternate Conditional Distributions Uncertanty as the Overlap of Alternate Condtonal Dstrbutons Olena Babak and Clayton V. Deutsch Centre for Computatonal Geostatstcs Department of Cvl & Envronmental Engneerng Unversty of Alberta An mportant

More information

A Bayes Algorithm for the Multitask Pattern Recognition Problem Direct Approach

A Bayes Algorithm for the Multitask Pattern Recognition Problem Direct Approach A Bayes Algorthm for the Multtask Pattern Recognton Problem Drect Approach Edward Puchala Wroclaw Unversty of Technology, Char of Systems and Computer etworks, Wybrzeze Wyspanskego 7, 50-370 Wroclaw, Poland

More information

Chapter 5 Multilevel Models

Chapter 5 Multilevel Models Chapter 5 Multlevel Models 5.1 Cross-sectonal multlevel models 5.1.1 Two-level models 5.1.2 Multple level models 5.1.3 Multple level modelng n other felds 5.2 Longtudnal multlevel models 5.2.1 Two-level

More information

Online Classification: Perceptron and Winnow

Online Classification: Perceptron and Winnow E0 370 Statstcal Learnng Theory Lecture 18 Nov 8, 011 Onlne Classfcaton: Perceptron and Wnnow Lecturer: Shvan Agarwal Scrbe: Shvan Agarwal 1 Introducton In ths lecture we wll start to study the onlne learnng

More information

Multiple Sound Source Location in 3D Space with a Synchronized Neural System

Multiple Sound Source Location in 3D Space with a Synchronized Neural System Multple Sound Source Locaton n D Space wth a Synchronzed Neural System Yum Takzawa and Atsush Fukasawa Insttute of Statstcal Mathematcs Research Organzaton of Informaton and Systems 0- Mdor-cho, Tachkawa,

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

International Journal of Mathematical Archive-3(3), 2012, Page: Available online through ISSN

International Journal of Mathematical Archive-3(3), 2012, Page: Available online through   ISSN Internatonal Journal of Mathematcal Archve-3(3), 2012, Page: 1136-1140 Avalable onlne through www.ma.nfo ISSN 2229 5046 ARITHMETIC OPERATIONS OF FOCAL ELEMENTS AND THEIR CORRESPONDING BASIC PROBABILITY

More information

Ensemble of GA based Selective Neural Network Ensembles

Ensemble of GA based Selective Neural Network Ensembles Ensemble of GA based Selectve eural etwork Ensembles Jan-Xn WU Zh-Hua ZHOU Zhao-Qan CHE atonal Laboratory for ovel Software Technology anjng Unversty anjng, 0093, P.R.Chna wujx@a.nju.edu.cn {zhouzh, chenzq}@nju.edu.cn

More information

829. An adaptive method for inertia force identification in cantilever under moving mass

829. An adaptive method for inertia force identification in cantilever under moving mass 89. An adaptve method for nerta force dentfcaton n cantlever under movng mass Qang Chen 1, Mnzhuo Wang, Hao Yan 3, Haonan Ye 4, Guola Yang 5 1,, 3, 4 Department of Control and System Engneerng, Nanng Unversty,

More information

Fuzzy Boundaries of Sample Selection Model

Fuzzy Boundaries of Sample Selection Model Proceedngs of the 9th WSES Internatonal Conference on ppled Mathematcs, Istanbul, Turkey, May 7-9, 006 (pp309-34) Fuzzy Boundares of Sample Selecton Model L. MUHMD SFIIH, NTON BDULBSH KMIL, M. T. BU OSMN

More information

Homework Assignment 3 Due in class, Thursday October 15

Homework Assignment 3 Due in class, Thursday October 15 Homework Assgnment 3 Due n class, Thursday October 15 SDS 383C Statstcal Modelng I 1 Rdge regresson and Lasso 1. Get the Prostrate cancer data from http://statweb.stanford.edu/~tbs/elemstatlearn/ datasets/prostate.data.

More information

Radial-Basis Function Networks

Radial-Basis Function Networks Radal-Bass uncton Networs v.0 March 00 Mchel Verleysen Radal-Bass uncton Networs - Radal-Bass uncton Networs p Orgn: Cover s theorem p Interpolaton problem p Regularzaton theory p Generalzed RBN p Unversal

More information

Kernels in Support Vector Machines. Based on lectures of Martin Law, University of Michigan

Kernels in Support Vector Machines. Based on lectures of Martin Law, University of Michigan Kernels n Support Vector Machnes Based on lectures of Martn Law, Unversty of Mchgan Non Lnear separable problems AND OR NOT() The XOR problem cannot be solved wth a perceptron. XOR Per Lug Martell - Systems

More information

Negative Binomial Regression

Negative Binomial Regression STATGRAPHICS Rev. 9/16/2013 Negatve Bnomal Regresson Summary... 1 Data Input... 3 Statstcal Model... 3 Analyss Summary... 4 Analyss Optons... 7 Plot of Ftted Model... 8 Observed Versus Predcted... 10 Predctons...

More information

Multilayer neural networks

Multilayer neural networks Lecture Multlayer neural networks Mlos Hauskrecht mlos@cs.ptt.edu 5329 Sennott Square Mdterm exam Mdterm Monday, March 2, 205 In-class (75 mnutes) closed book materal covered by February 25, 205 Multlayer

More information

SHORT-TERM POWER FORECASTING BY STATISTICAL METHODS FOR PHOTOVOLTAIC PLANTS IN SOUTH ITALY

SHORT-TERM POWER FORECASTING BY STATISTICAL METHODS FOR PHOTOVOLTAIC PLANTS IN SOUTH ITALY 4 th Imeko TC19 Symposum on Envronmental Instrumentaton and Measurements Protectng Envronment, Clmate Changes and Polluton Control June 3-4, 213, Lecce, Italy SHORT-TERM POWER FORECASTING BY STATISTICAL

More information

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm Desgn and Optmzaton of Fuzzy Controller for Inverse Pendulum System Usng Genetc Algorthm H. Mehraban A. Ashoor Unversty of Tehran Unversty of Tehran h.mehraban@ece.ut.ac.r a.ashoor@ece.ut.ac.r Abstract:

More information

Markov Chain Monte Carlo Lecture 6

Markov Chain Monte Carlo Lecture 6 where (x 1,..., x N ) X N, N s called the populaton sze, f(x) f (x) for at least one {1, 2,..., N}, and those dfferent from f(x) are called the tral dstrbutons n terms of mportance samplng. Dfferent ways

More information

Study on Active Micro-vibration Isolation System with Linear Motor Actuator. Gong-yu PAN, Wen-yan GU and Dong LI

Study on Active Micro-vibration Isolation System with Linear Motor Actuator. Gong-yu PAN, Wen-yan GU and Dong LI 2017 2nd Internatonal Conference on Electrcal and Electroncs: echnques and Applcatons (EEA 2017) ISBN: 978-1-60595-416-5 Study on Actve Mcro-vbraton Isolaton System wth Lnear Motor Actuator Gong-yu PAN,

More information

Credit Card Pricing and Impact of Adverse Selection

Credit Card Pricing and Impact of Adverse Selection Credt Card Prcng and Impact of Adverse Selecton Bo Huang and Lyn C. Thomas Unversty of Southampton Contents Background Aucton model of credt card solctaton - Errors n probablty of beng Good - Errors n

More information

Linear Feature Engineering 11

Linear Feature Engineering 11 Lnear Feature Engneerng 11 2 Least-Squares 2.1 Smple least-squares Consder the followng dataset. We have a bunch of nputs x and correspondng outputs y. The partcular values n ths dataset are x y 0.23 0.19

More information

Comparative Studies of Law of Conservation of Energy. and Law Clusters of Conservation of Generalized Energy

Comparative Studies of Law of Conservation of Energy. and Law Clusters of Conservation of Generalized Energy Comparatve Studes of Law of Conservaton of Energy and Law Clusters of Conservaton of Generalzed Energy No.3 of Comparatve Physcs Seres Papers Fu Yuhua (CNOOC Research Insttute, E-mal:fuyh1945@sna.com)

More information

Clock-Gating and Its Application to Low Power Design of Sequential Circuits

Clock-Gating and Its Application to Low Power Design of Sequential Circuits Clock-Gatng and Its Applcaton to Low Power Desgn of Sequental Crcuts ng WU Department of Electrcal Engneerng-Systems, Unversty of Southern Calforna Los Angeles, CA 989, USA, Phone: (23)74-448 Massoud PEDRAM

More information

RBF Neural Network Model Training by Unscented Kalman Filter and Its Application in Mechanical Fault Diagnosis

RBF Neural Network Model Training by Unscented Kalman Filter and Its Application in Mechanical Fault Diagnosis Appled Mechancs and Materals Submtted: 24-6-2 ISSN: 662-7482, Vols. 62-65, pp 2383-2386 Accepted: 24-6- do:.428/www.scentfc.net/amm.62-65.2383 Onlne: 24-8- 24 rans ech Publcatons, Swtzerland RBF Neural

More information

Application research on rough set -neural network in the fault diagnosis system of ball mill

Application research on rough set -neural network in the fault diagnosis system of ball mill Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(4):834-838 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 Applcaton research on rough set -neural network n the

More information

Psychology 282 Lecture #24 Outline Regression Diagnostics: Outliers

Psychology 282 Lecture #24 Outline Regression Diagnostics: Outliers Psychology 282 Lecture #24 Outlne Regresson Dagnostcs: Outlers In an earler lecture we studed the statstcal assumptons underlyng the regresson model, ncludng the followng ponts: Formal statement of assumptons.

More information

8 Derivation of Network Rate Equations from Single- Cell Conductance Equations

8 Derivation of Network Rate Equations from Single- Cell Conductance Equations Physcs 178/278 - Davd Klenfeld - Wnter 2015 8 Dervaton of Network Rate Equatons from Sngle- Cell Conductance Equatons We consder a network of many neurons, each of whch obeys a set of conductancebased,

More information

arxiv:cs.cv/ Jun 2000

arxiv:cs.cv/ Jun 2000 Correlaton over Decomposed Sgnals: A Non-Lnear Approach to Fast and Effectve Sequences Comparson Lucano da Fontoura Costa arxv:cs.cv/0006040 28 Jun 2000 Cybernetc Vson Research Group IFSC Unversty of São

More information

Chapter - 2. Distribution System Power Flow Analysis

Chapter - 2. Distribution System Power Flow Analysis Chapter - 2 Dstrbuton System Power Flow Analyss CHAPTER - 2 Radal Dstrbuton System Load Flow 2.1 Introducton Load flow s an mportant tool [66] for analyzng electrcal power system network performance. Load

More information

Statistics II Final Exam 26/6/18

Statistics II Final Exam 26/6/18 Statstcs II Fnal Exam 26/6/18 Academc Year 2017/18 Solutons Exam duraton: 2 h 30 mn 1. (3 ponts) A town hall s conductng a study to determne the amount of leftover food produced by the restaurants n the

More information

Study on Project Bidding Risk Evaluation Based on BP Neural Network Theory

Study on Project Bidding Risk Evaluation Based on BP Neural Network Theory 1192 Proceedngs of the 7th Internatonal Conference on Innovaton & Management Study on Proect Bddng Rs Evaluaton Based on BP Neural Networ Theory Wang Xnzheng 1, He Png 2 1 School of Cvl Engneerng, Nanyang

More information

An Evolutionary Method of Neural Network in System Identification

An Evolutionary Method of Neural Network in System Identification Internatonal Journal of Intellgent Informaton Systems 206; 5(5): 75-8 http://www.scencepublshnggroup.com//s do: 0.648/.s.2060505.4 ISSN: 2328-7675 (Prnt); ISSN: 2328-7683 (Onlne) An Evolutonary Method

More information

1. Inference on Regression Parameters a. Finding Mean, s.d and covariance amongst estimates. 2. Confidence Intervals and Working Hotelling Bands

1. Inference on Regression Parameters a. Finding Mean, s.d and covariance amongst estimates. 2. Confidence Intervals and Working Hotelling Bands Content. Inference on Regresson Parameters a. Fndng Mean, s.d and covarance amongst estmates.. Confdence Intervals and Workng Hotellng Bands 3. Cochran s Theorem 4. General Lnear Testng 5. Measures of

More information

Operating conditions of a mine fan under conditions of variable resistance

Operating conditions of a mine fan under conditions of variable resistance Paper No. 11 ISMS 216 Operatng condtons of a mne fan under condtons of varable resstance Zhang Ynghua a, Chen L a, b, Huang Zhan a, *, Gao Yukun a a State Key Laboratory of Hgh-Effcent Mnng and Safety

More information

1 Convex Optimization

1 Convex Optimization Convex Optmzaton We wll consder convex optmzaton problems. Namely, mnmzaton problems where the objectve s convex (we assume no constrants for now). Such problems often arse n machne learnng. For example,

More information

Copyright 2017 by Taylor Enterprises, Inc., All Rights Reserved. Adjusted Control Limits for U Charts. Dr. Wayne A. Taylor

Copyright 2017 by Taylor Enterprises, Inc., All Rights Reserved. Adjusted Control Limits for U Charts. Dr. Wayne A. Taylor Taylor Enterprses, Inc. Adjusted Control Lmts for U Charts Copyrght 207 by Taylor Enterprses, Inc., All Rghts Reserved. Adjusted Control Lmts for U Charts Dr. Wayne A. Taylor Abstract: U charts are used

More information

Chapter 13: Multiple Regression

Chapter 13: Multiple Regression Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

Basic Business Statistics, 10/e

Basic Business Statistics, 10/e Chapter 13 13-1 Basc Busness Statstcs 11 th Edton Chapter 13 Smple Lnear Regresson Basc Busness Statstcs, 11e 009 Prentce-Hall, Inc. Chap 13-1 Learnng Objectves In ths chapter, you learn: How to use regresson

More information

Introduction to the Introduction to Artificial Neural Network

Introduction to the Introduction to Artificial Neural Network Introducton to the Introducton to Artfcal Neural Netork Vuong Le th Hao Tang s sldes Part of the content of the sldes are from the Internet (possbly th modfcatons). The lecturer does not clam any onershp

More information

Determining Transmission Losses Penalty Factor Using Adaptive Neuro Fuzzy Inference System (ANFIS) For Economic Dispatch Application

Determining Transmission Losses Penalty Factor Using Adaptive Neuro Fuzzy Inference System (ANFIS) For Economic Dispatch Application 7 Determnng Transmsson Losses Penalty Factor Usng Adaptve Neuro Fuzzy Inference System (ANFIS) For Economc Dspatch Applcaton Rony Seto Wbowo Maurdh Hery Purnomo Dod Prastanto Electrcal Engneerng Department,

More information

A new Approach for Solving Linear Ordinary Differential Equations

A new Approach for Solving Linear Ordinary Differential Equations , ISSN 974-57X (Onlne), ISSN 974-5718 (Prnt), Vol. ; Issue No. 1; Year 14, Copyrght 13-14 by CESER PUBLICATIONS A new Approach for Solvng Lnear Ordnary Dfferental Equatons Fawz Abdelwahd Department of

More information