A Multimodal Fusion Algorithm Based on FRR and FAR Using SVM

Size: px
Start display at page:

Download "A Multimodal Fusion Algorithm Based on FRR and FAR Using SVM"

Transcription

1 Internatonal Journal of Securty and Its Applcatons A Multmodal Fuson Algorthm Based on FRR and FAR Usng SVM Yong L 1, Meme Sh 2, En Zhu 3, Janpng Yn 3, Janmn Zhao 4 1 Department of Informaton Engneerng, Engneerng Unversty of CAPF, X an Chna 2 Department of Scence, Engneerng Unversty of CAPF, X'an, Chna 3 School of Computer, Natonal Unversty of Defense Technology, Changsha, Hunan Provnce, Chna 4 Top Key Dscplne of Computer Software and Theory n Zhejang Provncal Colleges, Zhejang Normal Unversty, Chna 1 lyong@nudt.edu.cn, 2 memesh2003@yahoo.com.cn, 3 {enzhu,jpyn}@nudt.edu.cn, 4 zhaojm0310@163.com Abstract Remarable mprovements n recognton can be acheved through multbometrc fuson. Among varous fuson technques, score level fuson s the most frequently used n multbometrc system. In ths paper, we propose a novel fuson algorthm based on False Reject Rate (FRR) and False Accept Rate (FAR) usng Support Vector Machne (SVM). It transfers scores nto correspondng FRRs and FARs, thus avodng calculatng posteror probablty of a certan score, as well as be capable of llustratng dstrbuton of matchng scores. The proposed method taes full advantages of both capabltes of FRR and FAR to descrbe the order of score and classfcaton of SVM. Expermental results show that the proposed method outperforms exstng representatve approaches and can effectvely mprove the performance of multbometrc system. Keywords: Bometrcs, Multbometrc, Score level fuson, Mult-modal, SVM, False Reject Rate, False Accept Rate 1. Introducton Bometrc recognton refers to the process of automatcally confrmng the dentty of a person through physologcal or behavoral characterstcs. A hgh securty level system always demands for hgh performance, such as low FRR (False Reject Rate) and FAR (False Accept Rate). Unbometrc system, usng a sngle bometrc trat, wth ts nsuffcent nformaton and easly nfluenced by collecton noses, s no longer able to meet the requrements of hgh level system [1]. A possble soluton s multbometrc fuson [14-16]. Accordng to the level of nformaton fused, the fuson scheme can be classfed nto sensor level, feature level, score level and decson level fuson. Among them, score level fuson s the most frequently researched for t shelds both the dversty of bottom bometrc characterstc data and complexty of recognton process, as well as preserves smlarty metrcs of ndvdual trats to dfferentate samples from genune or ntruder. Already exstng score level fuson approaches can be categorzed nto three classes: transformaton-based, densty-based and classfer-based fuson [2]. Transformaton-based fuson frstly normalzes the matchng scores and then acqures a new score through a certan fuson rule to mae fnal decson. Ths method taes nto consderaton of the two factors, one s normalzaton functon, and the other s fuson rule [2, 65

2 Internatonal Journal of Securty and Its Applcatons 3-4]. In densty-based fuson scheme, matchng scores are frstly transformed nto posteror probablty and fnal decson s made accordng to Byes rules [5-6]. Classfer-based fuson taes N matchng scores as an N-dmensonal feature vector and transfers fuson problem nto classfcaton of the N-dmensonal vector [8-9]. In concluson, transformaton-based fuson technque, wth no tranng process and few consderaton of dstrbuton of matchng scores, s easy to mplement. On the contrary, densty based fuson, whch requres accurate estmaton of densty and huge number of tranng samples, s hard to carry out for the followng reasons: Frstly, postve samples, namely genune matchng scores are lmted n today's mult-bometrcs system. Secondly, t s dffcult to estmate the densty of matchng scores n that they may not obey a certan dstrbuton model. Besdes, the tme and space cost of ths method s consderable. Classfer based fuson has ts advantage of freeng from the restrcton of dfferent dstrbuton of matchng scores. However, t can nether drectly produce a matchng score nor obtan optmal FRR by certan FAR. Also, t needs suffcent tranng samples and new tranng for a dfferent system. Based on the above analyss of the advantages and dsadvantages of exstng methods, we propose a novel approach to descrbe the dstrbuton of matchng scores by FRR and FAR. As two mportant parameters of llustratng performance of a recognton system, FRR and FAR avod estmaton of probablty densty when used as descrptor of matchng score dstrbuton. It s smlar to ntegral of matchng score densty, thus maes the orgnal dscrete densty contnuous. Tronc [7] utlzed FRR-FAR as an mportant standard to mae the fnal decson, but wth the assumpton of normal dstrbuton when calculatng ts value. However, no matter how FRR and FAR are combned, loss of nformaton s nevtable. Besdes, the recognton performance of sum rule s not good enough. In ths context, we consder drectly usng FRR and FAR as features. Wth more features ntroduced, avalable nformaton s much more abundant. It frstly calculates FRR and FAR based on tranng samples, then computes FRR and FAR of each matchng score through nterpolaton based on testng samples, and maes fnal decson by SVM. Compared wth exstng methods, the algorthm presented n ths paper can greatly mprove the recognton performance of multbometrc system. 2. Transform the Score to FRR and FAR Matchng scores comng from dfferent recognton systems are not comparable drectly. At the same tme, FRR and FAR can be obtaned by settng threshold as each score. They can well descrbe the score dstrbuton of genune and mposter. Therefore, we frstly calculate FRR and FAR of each matchng score based on tranng samples, then compute FRR and FAR of each matchng score through nterpolaton based on testng samples FRR and FAR calculaton based on tranng set Suppose an M-matcher fuson system has N tranng samples. For th tranng sample, the template feature s 1 M U u,..., u, nput feature s 1 M V v,..., u and output score s S 1 s,..., s M. Here, u, v, s respectvely represents template feature, nput feature and matchng score of th tranng sample by th matcher. In addton, we use z to note when th tranng sample s genune or mposter. Namely, f the th tranng sample s genune, then z =1; f t s mposter, then z = 0. Then, FRR and FAR of th (1 M ) matcher based on tranng samples are respectvely calculated as follows. 66

3 Internatonal Journal of Securty and Its Applcatons N 1 N 1 FAR ( t) s t, z 0 (1 z ) (1) FRR ( t) s t, z 1 z (2) 2.2. FRR and FAR calculaton based on testng set Accordng to the above analyss, calculaton of transformaton functon based on FRR and FAR requres all the matchng scores, whch s mpossble n a real system. So we can only mae the calculaton based on tranng samples. Generally, tranng and testng samples follow the same dstrbuton, whch means that f one score appears n the tranng samples, t wll probably appear n the testng samples as well. Thus, usng each matchng score as threshold t, we can get transformaton value. For the newly appearng scores, we use nterpolaton to obtan ts transformaton value. At most N dscrete matchng scores are generated by each matcher for N tranng samples n tranng set. Generally, score obtaned n testng process s not absolutely dentcal to that attaned n tranng process. So the transformatonal scores n testng process can be calculated through nterpolaton based on scores from tranng samples. Suppose n the tranng process, n( n N) dfferent matchng scores are generated by th matcher based on N samples and raned n ascendng order { t, t,..., t }. Substtute these scores nto FAR () t and FRR () t, we can obtan as 1 2 n { FAR ( t ) 1 n } and { FRR ( t ) 1 n }. For matchng score x generated by th matcher n testng process, FAR ( x ) and FRR ( x ) are calculated n the four stuatons as follows. 1. If x t1, then use t1 and t 2 as control ponts, and calculate FRR ( x) and FAR ( x) through extrapolaton. Suppose t2 x x t1 FRR ( x) FRR ( t1) FRR ( t2) (3) t t t t t x x t FAR ( x) FAR ( t ) FAR ( t ) t2 t1 t2 t1 Then FRR FRR 0 FRR (5) 0 FRR 0 FAR FAR 1 FAR (6) 1 FAR 1 2. If x t n, then use t n and t n 1 as control ponts, and calculate FRR ( x) and FAR ( x) through extrapolaton. Suppose tn x x t n1 FRR ( x) FRR ( tn ) FRR ( ) 1 t n (7) t t t t n n1 n n 1 (4) 67

4 Internatonal Journal of Securty and Its Applcatons Then t x x t FAR x FAR t FAR t n n1 ( ) ( n ) ( ) 1 n tn t n t 1 n t n 1 FRR FRR 1 FRR 1 FRR 1 FAR FAR 0 FAR 0 FAR 0 (8) (9) (10) 3. If there exsts whch maes x t, FRR ( x) FRR( t ), FAR ( x) FAR( t ) (11) 4. If there exsts whch maes t x t 1, then calculate FRR ( x) and FAR ( x ) through nterpolaton. t x x t FRR x FRR t FRR t 1 ( ) ( ) ( 1) t 1t t 1t (12) t x x t FAR x FAR t FAR t 1 ( ) ( ) ( 1) t 1t t 1t (13) In step (3) and (4), bnary searchng s adopted to fnd correspondng ( t, t 1( 1 n ) ). The tme complexty of bnary search s O(log 2 N ), thus effectvely reducng tme complexty of the whole algorthm. So far, we have attaned FRR and FAR for each matchng score n testng samples An example for computaton of FRR and FAR To better demonstrate how our algorthm wors, we gve an example n ths secton. Table 1 s one part of the experment based on tranng samples, where matchng scores are raned n ascendng order. From the above analyss, we now that computaton of matchng score transformaton of a certan smple classfer s rrelevant to other classfers. So we just choose one smple classfer for llustraton. Also, n the tranng set, there s no other matchng score among the exstng ones, and FRR and FAR are calculated by tang matchng score as threshold based on tranng samples. Table 2 s one part of the experment based on testng samples, where matchng scores are among those n Table 1. We use calculaton formula of FAR () t and FRR () t based on testng samples. Tae the fourth matchng score x= as example, as we havet 4 x t 5, then t5 x xt4 FRR ( x) FRR ( t4) FRR ( t5) (14) t t t t t x xt FAR ( x) FAR ( t ) FAR ( t ) t5 t4 t5 t4 (15) 68

5 Internatonal Journal of Securty and Its Applcatons Table 1. Computaton of FRR and FAR on tranng set t FRR ( t ) FAR ( t ) Table 2. Computaton of FRR and FAR on testng set x FRR (x) FAR (x) SVM Fuson Based on FRR and FAR Durng score level fuson, score vector s combned to generate a sngle scalar score whch s later used to mae the fnal decson. Based on Secton 2, one match score can be transformed to two scores whch are ts FRR and FAR. All the tranng and testng scores can be transformed nto ther FRRs and FARs. SVM s based on the structural rs mnmzaton prncple [12], whch taes both the tranng error and generalzaton error nto consderaton. The SVM s to fnd an optmal hyper-plane that can well separate two class samples n the feature space. In our proposed multmodal bometrc method, we use SVM to buld a fuson functon whch can provde a fused score. Assume that we have N feature vectors wth assocated label y { 1, 1}, so the tranng data set s denoted as N n {( X, y )} X R, 1,2, N (16) After transformng the match scores nto FRRs and FARs, SVM s used to generate the fnal score. And X FRR ( s ), FAR ( s ), FRR ( s ), FAR ( s ),..., FRR ( s ), FAR ( s ) (17) M M Also, the score can be frst transformed by normalzed methods such as Tanh, MnMax as well as T, and then SVM s used (respectvely denoted as Tanh_SVM, MM_SVM, and T_SVM). In the above process, the sequental mnmal optmzaton (SMO) tranng algorthm [11] s used to enhance the tme effcency. In general, the dscrmnate functon of 69

6 Internatonal Journal of Securty and Its Applcatons an arbtrary classfer s of lttle probablty meanng. However, the probablstc output of SVM classfer can help n multmodal fuson post-processng. Fttng a sgmod functon to the classfer output s a soluton to ths problem. Platt [11] has descrbed a method of fttng the sgmod functon. 4. Experment Results The XM2VTS-Benchmar database [10] conssts of fve face matchers and three speech matchers and was parttoned nto tranng and evaluaton sets accordng to the Lausanne Protocol-1 (LPI). The benchmar of LPI ncludes two fles, one s dev.label and the other s eva.label. We use dev.label as tranng data and eva.label as test data. Our experments are conducted based on ths match score benchmar. We sgn the face matcher as F1, F2, F3, F4 and F5 and the speech matcher as S1, S2 and S3 respectvely. The Equal Error Rate (EER) s the pont of the ROC curve where the two errors,.e., the FRR and the FAR, are equal. Ths performance measure s wdely used n the bometrc feld to assess the performance of bometrc systems [2]. In bometrc recognton systems, we always try to mae EER smaller. 4.1 Comparson wth dfferent fuson strateges We frstly conducted experments to measure the advantages of the proposed method FF_SVM over densty-based algorthm LR [6], transformaton-based fuson rules: Tanh and MnMax wth smple sum rule [2] and the fuson methods T whch was proposed n Ref.[11]. The EER of all the matchers can be found n Table 3. As shown n Table 3, among face matchers, matcher face-3 and face-5 gan the best and worst performance respectvely. And among speech matchers, the performance order s speech-1, speech-3 and speech-2. Table 3. EERs of each smple recognton system on XM2VTS-Benchmar S1 S2 S3 F1 F2 F3 F4 F5 EER Table 4. EER comparson of dfferent fuson technques EER(%) F1S1 F1S2 F1S3 F2S1 F2S2 F2S3 F3S1 F3S2 F3S3 F4S1 F4S2 F4S3 F5S1 F5S2 F5S3 FF_SVM T LR Tanh MM The experments are conducted wth 15 nd multmodal combnaton. Table 2 shows the EER of mult-modal fuson among FF_SVM, LR, T, Tanh and MnMax. From Table 4, we observe that FF_SVM provdes the best performance 8 tmes and the second best performance 6 tmes. FF_SVM fuson method gans the best recognton performance followed by T and LR fuson algorthms. In order to evaluate the performance precsely, we gve each matcher of every fuson a performance mar. The performance mar for the best matcher s 5 and followed s 4, 3, 2, 1. If the performance of two matchers are the same, for 70

7 Internatonal Journal of Securty and Its Applcatons example, both are the second best, then the two matchers get the same mar (4+3)/2=3.5. Table 5 s the performance mar of dfferent fuson technques whch s measured by EER. From Table 5, we can easly fnd that the proposed fuson method FF_SVM shows the best performance because the total mar s the largest one. And we observe that the T and LR methods also perform well. Table 5. Performance mar comparson of dfferent technques based on EER Mar F1S1 F1S2 F1S3 F2S1 F2S2 F2S3 F3S1 F3S2 F3S3 F4S1 F4S2 F4S3 F5S1 F5S2 F5S3 total FF_SVM T LR Tanh MM Comparson wth SVM-based strateges In order to verfy the effect of transformng score to FRR and FAR, we compare the proposed method FF_SVM wth T-based SVM fuson method T_SVM, Tanh-based SVM fuson method Tanh_SVM, MnMax-based SVM fuson method (MM_SVM) and SVM fuson drectly on raw scores. From Table 6 and 7, we can observe that the proposed fuson method FF_SVM outperforms Tanh_SVM, MM_SVM, T_SVM and SVM. Durng 15 multmodal combnatons, FF_SVM gves the best performance 10 tmes, and the total performance mar s larger than the other SVM-based algorthms. Table 6. EER comparson of SVM-based Technques EER(%) F1S1 F1S2 F1S3 F2S1 F2S2 F2S3 F3S1 F3S2 F3S3 F4S1 F4S2 F4S3 F5S1 F5S2 F5S3 FF_SVM T_SVM Tanh_SVM MM_SVM SVM Table 7. Performance mar comparson of SVM-based Technques based on EER Mar F1S1 F1S2 F1S3 F2S1 F2S2 F2S3 F3S1 F3S2 F3S3 F4S1 F4S2 F4S3 F5S1 F5S2 F5S3 total FF_SVM T_SVM Tanh_SVM MM_SVM SVM To show the comparson of all the algorthms n multmodal bometrc systems, Fgure 1 shows the EERs of the 9 algorthms. The proposed FF_SVM algorthm s show n bold lne. From the fgure, we can conclude that the proposed algorthm can mprove the performance of multmodal bometrc system effcently. 71

8 EER Internatonal Journal of Securty and Its Applcatons EER-MM EER-MM-SVM EER-LR EER-SVM EER-T EER-T-SVM EER-Tanh EER-Tanh-SVM EER-FF-SVM Concluson Fgure 1. EERs of all Algorthms Ths paper proposes a novel fuson algorthm: FF_SVM. Ths proposed algorthm frst transforms the match scores nto FRRs and FARs and then adopts SVM for classfcaton. To be more specfc, we frstly calculate FRRs and FARs based on the tranng set, and then use nterpolaton to obtan the FRRs and FARs of each score n the testng set. As well, bnary searchng s appled to fnd nterpolaton ponts n order to reduce tme complexty of the whole algorthm. In addton, SVM s used to explore the relatonshp between the fnal score and FRR & FAR. The expermental results show that ths FF_SVM fuson algorthm can effcently mprove performance compared wth LR, T, Tanh and MnMax algorthms. Besdes, n comparson wth other SVM-based fuson methods, the proposed fuson method can provde better recognton performance for multmodal bometrc fuson systems. Acnowledgments Ths wor was supported by the Natonal Natural Scence Foundaton of Chna (Grant no , , ), the Foundaton for the Author of Natonal Excellent Doctoral Dssertaton (Grant No. 2007B4), and the Openng Fund of Top Key Dscplne of Computer Software and Theory n Zhejang Provncal Colleges at Zhejang Normal Unversty(Grant No. ZSDZZZZXK38). References [1] A. K. Jan, Bometrc Recognton: Overvew and Recent Advances, CIARP, LNCS 4756, (2007), pp [2] A. Ross, K. Nandaumar and A. K. Jan, Handboo of Multbometrcs, Sprnger Verlag, (2006). [3] M. Indovna, U. Uludag, R. Snelc, A. Mn and A. Jan, Multmodal Bometrc Authentcaton Methods: A COTS Approach, Proceedngs of Mult-Modal User Authentcaton (MMUA), (2003), pp [4] M. X. He and S. J. Horng, Performance evaluaton of score level fuson n multmodal bometrc systems, Pattern Recognton, vol. 43, no. 5, (2010), pp [5] A. K. Jan, K. Nandaumar and A. Ross, Score Normalzaton n Multmodal Bometrc Systems, Pattern Recognton, vol. 38, no. 12, (2005), pp

9 Internatonal Journal of Securty and Its Applcatons [6] K. Nandaumar and Y. Chen, Lelhood rato-based bometrc score fuson, IEEE Transactons on Pattern Analyss and Machne Intellgence, vol. 30, no. 2, (2008), pp [7] G. Gacnto, F. Rol and R. Tronc, Score Selecton Technques for Fngerprnt Mult-modal Bometrc Authentcaton, Proc. of 13th ICIAP, (2005), pp [8] A. Kumar and V. Kanhangad, A New Framewor for Adaptve Multmodal Bometrcs Management, IEEE Transactons on Informaton Forenscs and Securty, vol. 5, no. 2, (2010), pp [9] R. Tronc and G. Gacnto, Dynamc score combnaton of bnary experts, 19th Internatonal Conference on Pattern Recognton, New Yor, vol. 1-6, (2008), pp [10] N. Poh and S. Bengo, Database, Protocol and Tools for Evaluatng Score-Level Fuson Algorthms n Bometrc Authentcaton, Pattern Recognton, vol. 39, no. 2, (2006), pp [11] Y. L, J. Yn, J. Long and E. Zhu, A Novel Method for Multbometrc Fuson Based on FAR and FRR, MDAI 2009, LNAI 5861, (2009), pp [12] V. Vapn, The nature of statstcal learnng theory, Sprnger-Verlag, NewYor, (1995). [13] J. C. Platt, Probablstc Outputs for Support Vector Machnes for Pattern Recognton, U.Fayyad, Kluwer Academc Publshers: Boston, (1999). [14] D. Kumar and Y. Ryu, A Bref Introducton of bometrcs and Fngerprnt Payment Technology, Internatonal Journal of Advanced Scence and Technology, vol. 4, (2009), pp [15] M. Soltane, N. Doghmane and N. Guers, Face and Speech Based Mult-Modal Bometrc Authentcaton, Internatonal Journal of Advanced Scence and Technology, vol. 21, (2010), pp [16] D. Ranjan Ksu, P. Gupta and J. KantaSng, Multbometrcs Feature Level Fuson by Graph Clusterng, Internatonal Journal of Advanced Scence and Technology, vol. 5, (2011), pp Authors Yong L receved hs PhD degree from Natonal Unversty of Defense Technology of Chna. And now he s a lecturer at the Department of Informaton Engneerng, Engneerng Unversty of CAPF. Meme Sh receved her M.S. degree from Hunan Unversty of Chna. And now she s a lecturer of Engneerng Unversty of CAPF. En Zhu s an assocate professor at the School of Computer, Natonal Unversty of Defense Technology. Hs research nterest covers pattern recognton and mage process. 73

10 Internatonal Journal of Securty and Its Applcatons Janpng Yn s a professor at the School of Computer, Natonal Unversty of Defense Technology. Hs re-search nterest covers pattern recognton, artfcal ntellgence, and networ securty. Janmn Zhao s a professor of Zhejang Normal Unversty. Hs research nterest covers nformaton securty, pattern recognton and mage process. 74

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

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

An Improved multiple fractal algorithm

An Improved multiple fractal algorithm Advanced Scence and Technology Letters Vol.31 (MulGraB 213), pp.184-188 http://dx.do.org/1.1427/astl.213.31.41 An Improved multple fractal algorthm Yun Ln, Xaochu Xu, Jnfeng Pang College of Informaton

More information

A Particle Filter Algorithm based on Mixing of Prior probability density and UKF as Generate Importance Function

A Particle Filter Algorithm based on Mixing of Prior probability density and UKF as Generate Importance Function Advanced Scence and Technology Letters, pp.83-87 http://dx.do.org/10.14257/astl.2014.53.20 A Partcle Flter Algorthm based on Mxng of Pror probablty densty and UKF as Generate Importance Functon Lu Lu 1,1,

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

A Network Intrusion Detection Method Based on Improved K-means Algorithm

A Network Intrusion Detection Method Based on Improved K-means Algorithm Advanced Scence and Technology Letters, pp.429-433 http://dx.do.org/10.14257/astl.2014.53.89 A Network Intruson Detecton Method Based on Improved K-means Algorthm Meng Gao 1,1, Nhong Wang 1, 1 Informaton

More information

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS HCMC Unversty of Pedagogy Thong Nguyen Huu et al. A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS Thong Nguyen Huu and Hao Tran Van Department of mathematcs-nformaton,

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

Semi-supervised Classification with Active Query Selection

Semi-supervised Classification with Active Query Selection Sem-supervsed Classfcaton wth Actve Query Selecton Jao Wang and Swe Luo School of Computer and Informaton Technology, Beng Jaotong Unversty, Beng 00044, Chna Wangjao088@63.com Abstract. Labeled samples

More information

Lecture 12: Classification

Lecture 12: Classification Lecture : Classfcaton g Dscrmnant functons g The optmal Bayes classfer g Quadratc classfers g Eucldean and Mahalanobs metrcs g K Nearest Neghbor Classfers Intellgent Sensor Systems Rcardo Guterrez-Osuna

More information

Statistical machine learning and its application to neonatal seizure detection

Statistical machine learning and its application to neonatal seizure detection 19/Oct/2009 Statstcal machne learnng and ts applcaton to neonatal sezure detecton Presented by Andry Temko Department of Electrcal and Electronc Engneerng Page 2 of 42 A. Temko, Statstcal Machne Learnng

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

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

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

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

MULTISPECTRAL IMAGE CLASSIFICATION USING BACK-PROPAGATION NEURAL NETWORK IN PCA DOMAIN

MULTISPECTRAL IMAGE CLASSIFICATION USING BACK-PROPAGATION NEURAL NETWORK IN PCA DOMAIN MULTISPECTRAL IMAGE CLASSIFICATION USING BACK-PROPAGATION NEURAL NETWORK IN PCA DOMAIN S. Chtwong, S. Wtthayapradt, S. Intajag, and F. Cheevasuvt Faculty of Engneerng, Kng Mongkut s Insttute of Technology

More information

Suppose that there s a measured wndow of data fff k () ; :::; ff k g of a sze w, measured dscretely wth varable dscretzaton step. It s convenent to pl

Suppose that there s a measured wndow of data fff k () ; :::; ff k g of a sze w, measured dscretely wth varable dscretzaton step. It s convenent to pl RECURSIVE SPLINE INTERPOLATION METHOD FOR REAL TIME ENGINE CONTROL APPLICATIONS A. Stotsky Volvo Car Corporaton Engne Desgn and Development Dept. 97542, HA1N, SE- 405 31 Gothenburg Sweden. Emal: astotsky@volvocars.com

More information

VQ widely used in coding speech, image, and video

VQ widely used in coding speech, image, and video at Scalar quantzers are specal cases of vector quantzers (VQ): they are constraned to look at one sample at a tme (memoryless) VQ does not have such constrant better RD perfomance expected Source codng

More information

Microwave Diversity Imaging Compression Using Bioinspired

Microwave Diversity Imaging Compression Using Bioinspired Mcrowave Dversty Imagng Compresson Usng Bonspred Neural Networks Youwe Yuan 1, Yong L 1, Wele Xu 1, Janghong Yu * 1 School of Computer Scence and Technology, Hangzhou Danz Unversty, Hangzhou, Zhejang,

More information

Unified Subspace Analysis for Face Recognition

Unified Subspace Analysis for Face Recognition Unfed Subspace Analyss for Face Recognton Xaogang Wang and Xaoou Tang Department of Informaton Engneerng The Chnese Unversty of Hong Kong Shatn, Hong Kong {xgwang, xtang}@e.cuhk.edu.hk Abstract PCA, LDA

More information

The Order Relation and Trace Inequalities for. Hermitian Operators

The Order Relation and Trace Inequalities for. Hermitian Operators Internatonal Mathematcal Forum, Vol 3, 08, no, 507-57 HIKARI Ltd, wwwm-hkarcom https://doorg/0988/mf088055 The Order Relaton and Trace Inequaltes for Hermtan Operators Y Huang School of Informaton Scence

More information

Natural Language Processing and Information Retrieval

Natural Language Processing and Information Retrieval Natural Language Processng and Informaton Retreval Support Vector Machnes Alessandro Moschtt Department of nformaton and communcaton technology Unversty of Trento Emal: moschtt@ds.untn.t Summary Support

More information

Using T.O.M to Estimate Parameter of distributions that have not Single Exponential Family

Using T.O.M to Estimate Parameter of distributions that have not Single Exponential Family IOSR Journal of Mathematcs IOSR-JM) ISSN: 2278-5728. Volume 3, Issue 3 Sep-Oct. 202), PP 44-48 www.osrjournals.org Usng T.O.M to Estmate Parameter of dstrbutons that have not Sngle Exponental Famly Jubran

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

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

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

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

Discretization of Continuous Attributes in Rough Set Theory and Its Application*

Discretization of Continuous Attributes in Rough Set Theory and Its Application* Dscretzaton of Contnuous Attrbutes n Rough Set Theory and Its Applcaton* Gexang Zhang 1,2, Lazhao Hu 1, and Wedong Jn 2 1 Natonal EW Laboratory, Chengdu 610036 Schuan, Chna dylan7237@sna.com 2 School of

More information

Neural networks. Nuno Vasconcelos ECE Department, UCSD

Neural networks. Nuno Vasconcelos ECE Department, UCSD Neural networs Nuno Vasconcelos ECE Department, UCSD Classfcaton a classfcaton problem has two types of varables e.g. X - vector of observatons (features) n the world Y - state (class) of the world x X

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

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number

More information

Power law and dimension of the maximum value for belief distribution with the max Deng entropy

Power law and dimension of the maximum value for belief distribution with the max Deng entropy Power law and dmenson of the maxmum value for belef dstrbuton wth the max Deng entropy Bngy Kang a, a College of Informaton Engneerng, Northwest A&F Unversty, Yanglng, Shaanx, 712100, Chna. Abstract Deng

More information

CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING INTRODUCTION

CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING INTRODUCTION CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING N. Phanthuna 1,2, F. Cheevasuvt 2 and S. Chtwong 2 1 Department of Electrcal Engneerng, Faculty of Engneerng Rajamangala

More information

Orientation Model of Elite Education and Mass Education

Orientation Model of Elite Education and Mass Education Proceedngs of the 8th Internatonal Conference on Innovaton & Management 723 Orentaton Model of Elte Educaton and Mass Educaton Ye Peng Huanggang Normal Unversty, Huanggang, P.R.Chna, 438 (E-mal: yepeng@hgnc.edu.cn)

More information

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011 Stanford Unversty CS359G: Graph Parttonng and Expanders Handout 4 Luca Trevsan January 3, 0 Lecture 4 In whch we prove the dffcult drecton of Cheeger s nequalty. As n the past lectures, consder an undrected

More information

Uncertainty and auto-correlation in. Measurement

Uncertainty and auto-correlation in. Measurement Uncertanty and auto-correlaton n arxv:1707.03276v2 [physcs.data-an] 30 Dec 2017 Measurement Markus Schebl Federal Offce of Metrology and Surveyng (BEV), 1160 Venna, Austra E-mal: markus.schebl@bev.gv.at

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

x = , so that calculated

x = , so that calculated Stat 4, secton Sngle Factor ANOVA notes by Tm Plachowsk n chapter 8 we conducted hypothess tests n whch we compared a sngle sample s mean or proporton to some hypotheszed value Chapter 9 expanded ths to

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

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

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

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

DETERMINATION OF UNCERTAINTY ASSOCIATED WITH QUANTIZATION ERRORS USING THE BAYESIAN APPROACH

DETERMINATION OF UNCERTAINTY ASSOCIATED WITH QUANTIZATION ERRORS USING THE BAYESIAN APPROACH Proceedngs, XVII IMEKO World Congress, June 7, 3, Dubrovn, Croata Proceedngs, XVII IMEKO World Congress, June 7, 3, Dubrovn, Croata TC XVII IMEKO World Congress Metrology n the 3rd Mllennum June 7, 3,

More information

A New Scrambling Evaluation Scheme based on Spatial Distribution Entropy and Centroid Difference of Bit-plane

A New Scrambling Evaluation Scheme based on Spatial Distribution Entropy and Centroid Difference of Bit-plane A New Scramblng Evaluaton Scheme based on Spatal Dstrbuton Entropy and Centrod Dfference of Bt-plane Lang Zhao *, Avshek Adhkar Kouch Sakura * * Graduate School of Informaton Scence and Electrcal Engneerng,

More information

CSC 411 / CSC D11 / CSC C11

CSC 411 / CSC D11 / CSC C11 18 Boostng s a general strategy for learnng classfers by combnng smpler ones. The dea of boostng s to take a weak classfer that s, any classfer that wll do at least slghtly better than chance and use t

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

Sparse Gaussian Processes Using Backward Elimination

Sparse Gaussian Processes Using Backward Elimination Sparse Gaussan Processes Usng Backward Elmnaton Lefeng Bo, Lng Wang, and Lcheng Jao Insttute of Intellgent Informaton Processng and Natonal Key Laboratory for Radar Sgnal Processng, Xdan Unversty, X an

More information

Rotation Invariant Shape Contexts based on Feature-space Fourier Transformation

Rotation Invariant Shape Contexts based on Feature-space Fourier Transformation Fourth Internatonal Conference on Image and Graphcs Rotaton Invarant Shape Contexts based on Feature-space Fourer Transformaton Su Yang 1, Yuanyuan Wang Dept of Computer Scence and Engneerng, Fudan Unversty,

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

Feature Selection in Multi-instance Learning

Feature Selection in Multi-instance Learning The Nnth Internatonal Symposum on Operatons Research and Its Applcatons (ISORA 10) Chengdu-Juzhagou, Chna, August 19 23, 2010 Copyrght 2010 ORSC & APORC, pp. 462 469 Feature Selecton n Mult-nstance Learnng

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

Numerical Heat and Mass Transfer

Numerical Heat and Mass Transfer Master degree n Mechancal Engneerng Numercal Heat and Mass Transfer 06-Fnte-Dfference Method (One-dmensonal, steady state heat conducton) Fausto Arpno f.arpno@uncas.t Introducton Why we use models and

More information

Chapter 8 Indicator Variables

Chapter 8 Indicator Variables Chapter 8 Indcator Varables In general, e explanatory varables n any regresson analyss are assumed to be quanttatve n nature. For example, e varables lke temperature, dstance, age etc. are quanttatve n

More information

CHAPTER IV RESEARCH FINDING AND DISCUSSIONS

CHAPTER IV RESEARCH FINDING AND DISCUSSIONS CHAPTER IV RESEARCH FINDING AND DISCUSSIONS A. Descrpton of Research Fndng. The Implementaton of Learnng Havng ganed the whole needed data, the researcher then dd analyss whch refers to the statstcal data

More information

Automatic Object Trajectory- Based Motion Recognition Using Gaussian Mixture Models

Automatic Object Trajectory- Based Motion Recognition Using Gaussian Mixture Models Automatc Object Trajectory- Based Moton Recognton Usng Gaussan Mxture Models Fasal I. Bashr, Ashfaq A. Khokhar, Dan Schonfeld Electrcal and Computer Engneerng, Unversty of Illnos at Chcago. Chcago, IL,

More information

Statistical Energy Analysis for High Frequency Acoustic Analysis with LS-DYNA

Statistical Energy Analysis for High Frequency Acoustic Analysis with LS-DYNA 14 th Internatonal Users Conference Sesson: ALE-FSI Statstcal Energy Analyss for Hgh Frequency Acoustc Analyss wth Zhe Cu 1, Yun Huang 1, Mhamed Soul 2, Tayeb Zeguar 3 1 Lvermore Software Technology Corporaton

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

A Fast Computer Aided Design Method for Filters

A Fast Computer Aided Design Method for Filters 2017 Asa-Pacfc Engneerng and Technology Conference (APETC 2017) ISBN: 978-1-60595-443-1 A Fast Computer Aded Desgn Method for Flters Gang L ABSTRACT *Ths paper presents a fast computer aded desgn method

More information

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud Resource Allocaton wth a Budget Constrant for Computng Independent Tasks n the Cloud Wemng Sh and Bo Hong School of Electrcal and Computer Engneerng Georga Insttute of Technology, USA 2nd IEEE Internatonal

More information

Finding Dense Subgraphs in G(n, 1/2)

Finding Dense Subgraphs in G(n, 1/2) Fndng Dense Subgraphs n Gn, 1/ Atsh Das Sarma 1, Amt Deshpande, and Rav Kannan 1 Georga Insttute of Technology,atsh@cc.gatech.edu Mcrosoft Research-Bangalore,amtdesh,annan@mcrosoft.com Abstract. Fndng

More information

FORECASTING EXCHANGE RATE USING SUPPORT VECTOR MACHINES

FORECASTING EXCHANGE RATE USING SUPPORT VECTOR MACHINES Proceedngs of the Fourth Internatonal Conference on Machne Learnng and Cybernetcs, Guangzhou, 8- August 005 FORECASTING EXCHANGE RATE USING SUPPORT VECTOR MACHINES DING-ZHOU CAO, SU-LIN PANG, YUAN-HUAI

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

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

De-noising Method Based on Kernel Adaptive Filtering for Telemetry Vibration Signal of the Vehicle Test Kejun ZENG

De-noising Method Based on Kernel Adaptive Filtering for Telemetry Vibration Signal of the Vehicle Test Kejun ZENG 6th Internatonal Conference on Mechatroncs, Materals, Botechnology and Envronment (ICMMBE 6) De-nosng Method Based on Kernel Adaptve Flterng for elemetry Vbraton Sgnal of the Vehcle est Kejun ZEG PLA 955

More information

On an Extension of Stochastic Approximation EM Algorithm for Incomplete Data Problems. Vahid Tadayon 1

On an Extension of Stochastic Approximation EM Algorithm for Incomplete Data Problems. Vahid Tadayon 1 On an Extenson of Stochastc Approxmaton EM Algorthm for Incomplete Data Problems Vahd Tadayon Abstract: The Stochastc Approxmaton EM (SAEM algorthm, a varant stochastc approxmaton of EM, s a versatle tool

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

Speeding up Computation of Scalar Multiplication in Elliptic Curve Cryptosystem

Speeding up Computation of Scalar Multiplication in Elliptic Curve Cryptosystem H.K. Pathak et. al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng Speedng up Computaton of Scalar Multplcaton n Ellptc Curve Cryptosystem H. K. Pathak Manju Sangh S.o.S n Computer scence

More information

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1 Estmatng the Fundamental Matrx by Transformng Image Ponts n Projectve Space 1 Zhengyou Zhang and Charles Loop Mcrosoft Research, One Mcrosoft Way, Redmond, WA 98052, USA E-mal: fzhang,cloopg@mcrosoft.com

More information

Chapter 2 A Class of Robust Solution for Linear Bilevel Programming

Chapter 2 A Class of Robust Solution for Linear Bilevel Programming Chapter 2 A Class of Robust Soluton for Lnear Blevel Programmng Bo Lu, Bo L and Yan L Abstract Under the way of the centralzed decson-makng, the lnear b-level programmng (BLP) whose coeffcents are supposed

More information

Message modification, neutral bits and boomerangs

Message modification, neutral bits and boomerangs Message modfcaton, neutral bts and boomerangs From whch round should we start countng n SHA? Antone Joux DGA and Unversty of Versalles St-Quentn-en-Yvelnes France Jont work wth Thomas Peyrn 1 Dfferental

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

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

Pulse Coded Modulation

Pulse Coded Modulation Pulse Coded Modulaton PCM (Pulse Coded Modulaton) s a voce codng technque defned by the ITU-T G.711 standard and t s used n dgtal telephony to encode the voce sgnal. The frst step n the analog to dgtal

More information

Statistical Foundations of Pattern Recognition

Statistical Foundations of Pattern Recognition Statstcal Foundatons of Pattern Recognton Learnng Objectves Bayes Theorem Decson-mang Confdence factors Dscrmnants The connecton to neural nets Statstcal Foundatons of Pattern Recognton NDE measurement

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

MDL-Based Unsupervised Attribute Ranking

MDL-Based Unsupervised Attribute Ranking MDL-Based Unsupervsed Attrbute Rankng Zdravko Markov Computer Scence Department Central Connectcut State Unversty New Brtan, CT 06050, USA http://www.cs.ccsu.edu/~markov/ markovz@ccsu.edu MDL-Based Unsupervsed

More information

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling Real-Tme Systems Multprocessor schedulng Specfcaton Implementaton Verfcaton Multprocessor schedulng -- -- Global schedulng How are tasks assgned to processors? Statc assgnment The processor(s) used for

More information

Composite Hypotheses testing

Composite Hypotheses testing Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter

More information

Linear Approximation with Regularization and Moving Least Squares

Linear Approximation with Regularization and Moving Least Squares Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...

More information

Module 9. Lecture 6. Duality in Assignment Problems

Module 9. Lecture 6. Duality in Assignment Problems Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept

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

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

Probability Density Function Estimation by different Methods

Probability Density Function Estimation by different Methods EEE 739Q SPRIG 00 COURSE ASSIGMET REPORT Probablty Densty Functon Estmaton by dfferent Methods Vas Chandraant Rayar Abstract The am of the assgnment was to estmate the probablty densty functon (PDF of

More information

Other NN Models. Reinforcement learning (RL) Probabilistic neural networks

Other NN Models. Reinforcement learning (RL) Probabilistic neural networks Other NN Models Renforcement learnng (RL) Probablstc neural networks Support vector machne (SVM) Renforcement learnng g( (RL) Basc deas: Supervsed dlearnng: (delta rule, BP) Samples (x, f(x)) to learn

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

Finite Mixture Models and Expectation Maximization. Most slides are from: Dr. Mario Figueiredo, Dr. Anil Jain and Dr. Rong Jin

Finite Mixture Models and Expectation Maximization. Most slides are from: Dr. Mario Figueiredo, Dr. Anil Jain and Dr. Rong Jin Fnte Mxture Models and Expectaton Maxmzaton Most sldes are from: Dr. Maro Fgueredo, Dr. Anl Jan and Dr. Rong Jn Recall: The Supervsed Learnng Problem Gven a set of n samples X {(x, y )},,,n Chapter 3 of

More information

Large-Margin HMM Estimation for Speech Recognition

Large-Margin HMM Estimation for Speech Recognition Large-Margn HMM Estmaton for Speech Recognton Prof. Hu Jang Department of Computer Scence and Engneerng York Unversty, Toronto, Ont. M3J 1P3, CANADA Emal: hj@cs.yorku.ca Ths s a jont work wth Chao-Jun

More information

ADVANCED MACHINE LEARNING ADVANCED MACHINE LEARNING

ADVANCED MACHINE LEARNING ADVANCED MACHINE LEARNING 1 ADVANCED ACHINE LEARNING ADVANCED ACHINE LEARNING Non-lnear regresson technques 2 ADVANCED ACHINE LEARNING Regresson: Prncple N ap N-dm. nput x to a contnuous output y. Learn a functon of the type: N

More information

A fingerprint identification algorithm by clustering similarity

A fingerprint identification algorithm by clustering similarity Scence n Chna Ser. Informaton Scences 005 Vol.48 No.4 437 451 437 A fngerprnt dentfcaton algorthm by clusterng smlarty TIAN Je, HE Yulang, CHEN Hong & YAN Xn Center for Bometrc Research and Testng (CBRT),

More information

Recover plaintext attack to block ciphers

Recover plaintext attack to block ciphers Recover plantext attac to bloc cphers L An-Png Bejng 100085, P.R.Chna apl0001@sna.com Abstract In ths paper, we wll present an estmaton for the upper-bound of the amount of 16-bytes plantexts for Englsh

More information

Lecture 20: November 7

Lecture 20: November 7 0-725/36-725: Convex Optmzaton Fall 205 Lecturer: Ryan Tbshran Lecture 20: November 7 Scrbes: Varsha Chnnaobreddy, Joon Sk Km, Lngyao Zhang Note: LaTeX template courtesy of UC Berkeley EECS dept. Dsclamer:

More information

Learning Theory: Lecture Notes

Learning Theory: Lecture Notes Learnng Theory: Lecture Notes Lecturer: Kamalka Chaudhur Scrbe: Qush Wang October 27, 2012 1 The Agnostc PAC Model Recall that one of the constrants of the PAC model s that the data dstrbuton has to be

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

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

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 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

Keyword Reduction for Text Categorization using Neighborhood Rough Sets

Keyword Reduction for Text Categorization using Neighborhood Rough Sets IJCSI Internatonal Journal of Computer Scence Issues, Volume 1, Issue 1, No, January 015 ISSN (rnt): 1694-0814 ISSN (Onlne): 1694-0784 www.ijcsi.org Keyword Reducton for Text Categorzaton usng Neghborhood

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

GEMINI GEneric Multimedia INdexIng

GEMINI GEneric Multimedia INdexIng GEMINI GEnerc Multmeda INdexIng Last lecture, LSH http://www.mt.edu/~andon/lsh/ Is there another possble soluton? Do we need to perform ANN? 1 GEnerc Multmeda INdexIng dstance measure Sub-pattern Match

More information

Study of Selective Ensemble Learning Methods Based on Support Vector Machine

Study of Selective Ensemble Learning Methods Based on Support Vector Machine Avalable onlne at www.scencedrect.com Physcs Proceda 33 (2012 ) 1518 1525 2012 Internatonal Conference on Medcal Physcs and Bomedcal Engneerng Study of Selectve Ensemble Learnng Methods Based on Support

More information