Integrating Neural Networks and PCA for Fast Covert Surveillance

Size: px
Start display at page:

Download "Integrating Neural Networks and PCA for Fast Covert Surveillance"

Transcription

1 Integratng Neural Networks and PCA for Fast Covert Survellance Hazem M. El-Bakry, and Mamoon H. Mamoon Faculty of Computer Scence & Informaton Systems, Mansoura Unversty, EGYPT E-mal: Abstract In ths paper, a fast algorthm for covert survellance s presented. Such algorthm uses fast neural networks (FNNs) for human face detecton. The proposed FNNs apples cross correlaton n the frequency doman between the nput mage and the weghts of neural networks. For the purpose of effcent observaton and montorng by hgher admnstraton, we need to transfer the detected face to hgher admnstraton. Therefore, effcent compresson algorthm s needed. Here, the hybrd method k-pca s used. It s proved mathematcally and practcally that the proposed combned algorthms are effcent for face detecton and compresson. Keywords: Image Compresson, Vector Quantzaton, k-pca, Cross Correlaton, Frequency Doman, Covert Survellance I. Introducton Bometrcs offers new solutons for personal dentfcaton. Such solutons are user frendly and allow secure access. Bometrcs works by nspectng bologcal features of ourselves that dstngush us from others. Examples nclude fngerprnts, hand geometry, and voce recognton. It would be preferable to base recognton decsons upon features whch had very lttle genetc varatons, and stable over lfetme of the ndvduals. Among dfferent bometrcs technques, face recognton s the only sutable method for polce covert survellance. The human face s a complex pattern. Fndng human faces automatcally n a scene s a dffcult yet sgnfcant problem. It s the frst step n fully automatc human face recognton system. Face detecton s the fundamental step before the face recognton or dentfcaton procedure. Its relablty and tme response have a major nfluence on the performance and usablty of the whole face recognton system. For web ndexaton applcatons, the processng tme must be kept as low as possble as the number of mages on the web ncreases contnuously [19]. Among other technques [], neural networks are effcent face detectors [19-1]. admnstraton for montorng. Therefore, an effcent compresson algorthm s needed. In recent years, prncpal component analyss (PCA) has attracted great attenton n mage compresson. However, snce the compressed mage data nclude both the transformaton matrx (the egenvectors) and the transformed coeffcents, PCA cannot produce the performance lke DCT (dscrete cosne transform) n respect of compresson rato. In usng DCT, we need only to preserve the coeffcents after transformaton, because the transformaton matrx s unversal n the sense that t can be used to compress all mages. The authors [9-1] buld a unversal PCA by proposng a hybrd method called k-pca. The basc dea s to construct k sets of egenvectors for dfferent mage blocks wth dstnct characterstcs usng some tranng data. The k sets of egenvectors are then used to compress all mages. Vector quantzaton (VQ) s adopted to splt the tranng data space. Ths paper s organzed as follows: fast face detecton by usng hgh speed neural networks s gven n secton. In secton 3, the k-pca approach by combnng VQ and PCA s dscussed. Concluson s gven n secton 4. After face recognton, the detected face s requred to be transferred through computer networks to hgher. Fast Face Detecton by usng Hgh Speed Neural Networks ISSN: ISBN:

2 Frst neural networks are traned to classfy sub-mages whch contan face from those whch do not and ths s done n the spatal doman. In the test phase, each submage n the nput mage (under test) s tested for the presence or absence of human faces. At each pxel poston n the nput mage each sub-mage s multpled by a wndow of weghts, whch has the same sze as the sub-mage. Ths multplcaton s done n the spatal doman. The outputs of neurons n the hdden layer are multpled by the weghts of the output layer. When the fnal output s hgh ths means that the sub-mage under test contan human moton and vce versa. Thus, we may conclude that ths searchng problem s cross correlaton n the spatal doman between the mage under test and the nput weghts of neural networks. In ths secton, a fast algorthm for detectng human moton based on two dmensonal cross correlatons that take place between the tested mage and the sldng wndow (0x0 pxels) s descrbed. Such wndow s represented by the neural network weghts stuated between the nput unt and the hdden layer. The convoluton theorem n mathematcal analyss says that a convoluton of f wth h s dentcal to the result of the followng steps: let F and H be the results of the Fourer transformaton of f and h n the frequency doman. Multply F and H n the frequency doman pont by pont and then transform ths product nto spatal doman va the nverse Fourer transform [14-18]. As a result, these cross correlatons can be represented by a product n the frequency doman. Thus, by usng cross correlaton n the frequency doman a speed up n an order of magntude can be acheved durng the detecton process [14-18]. In the detecton phase, a sub-mage X of sze mxz (sldng wndow) s extracted from the tested mage, whch has a sze PxT, and fed to the neural network. Let W be the vector of weghts between the nput submage and the hdden layer. Ths vector has a sze of mxz and can be represented as mxz matrx. The output of hdden neurons h can be calculated as follows: z m h = g = W (j,k)x(j,k) + b j 1k = 1 (1) where g s the actvaton functon and b s the bas of each hdden neuron (). Eq.1 represents the output of each hdden neuron for a partcular sub-mage I. It can be computed for the whole mage Ψ as follows: h (u, v) = m/ z/ g W (j,k)ψ(u + j,v+ k) + b j = m/k = z/ () Eq. represents a cross correlaton operaton. Gven any two functons f and g, ther cross correlaton can be obtaned by []: g(x,y) f(x,y) = g(m,z)f(x+ m,y + z ) m= z= Therefore, Eq.3 can be wrtten as follows [14-19]: ( Ψ ) (3) h = g W + b (4) where h s the output of the hdden neuron () and h (u,v) s the actvty of the hdden unt () when the sldng wndow s located at poston (u,v) n the nput mage Ψ and (u,v) [P-m+1,T-n+1]. Now, the above cross correlaton can be expressed n terms of the Fourer Transform [14-19]: W Ψ = F 1 ( F( Ψ ) F* ( ) W (5) (*) means the conjugate of the FFT for the weght matrx. Hence, by evaluatng ths cross correlaton, a speed up rato can be obtaned comparable to conventonal neural networks. Also, the fnal output of the neural network can be evaluated as follows [14-19]: q O(u, v) = g = Wo() h (u,v) + bo (6) 1 where q s the number of neurons n the hdden layer. O(u,v) s the output of the neural network when the sldng wndow located at the poston (u,v) n the nput mage Ψ. W o s the weght matrx between hdden and output layer. The complexty of cross correlaton n the frequency doman can be analyzed as follows: 1. For a tested mage of NxN pxels, the D-FFT requres a number equal to N log N of complex computaton steps. Also, the same number of complex computaton steps s requred for computng the D- ISSN: ISBN:

3 FFT of the weght matrx for each neuron n the hdden layer.. At each neuron n the hdden layer, the nverse D- FFT s computed. So, q backward and (1+q) forward transforms have to be computed. Therefore, for an mage under test, the total number of the D-FFT to compute s (q+1)n log N. 3. The nput mage and the weghts should be multpled n the frequency doman. Therefore, a number of complex computaton steps equal to qn should be added. 4. The number of computaton steps requred by the fast neural networks s complex and must be converted nto a real verson. It s known that the two dmensonal Fast Fourer Transform requres (N /)log N complex multplcatons and N log N complex addtons [13]. Every complex multplcaton s realzed by sx real floatng pont operatons and every complex addton s mplemented by two real floatng pont operatons. So, the total number of computaton steps requred to obtan the D-FFT of an NxN mage s: whch may be smplfed to: ρ=6((n /)log N ) + (N log N ) (7) ρ=5n log N (8) Performng complex dot product n the frequency doman also requres 6qN real operatons. 5. In order to perform cross correlaton n the frequency doman, the weght matrx must have the same sze as the nput mage. Assume that the nput object has a sze of (nxn) dmensons. So, the search process wll be done over sub-mages of (nxn) dmensons and the weght matrx wll have the same sze. Therefore, a number of zeros = (N -n ) must be added to the weght matrx. Ths requres a total real number of computaton steps = q(n -n ) for all neurons. Moreover, after computng the D-FFT for the weght matrx, the conjugate of ths matrx must be obtaned. So, a real number of computaton steps =qn should be added n order to obtan the conjugate of the weght matrx for all neurons. Also, a number of real computaton steps equal to N s requred to create butterfles complex numbers (e -jk(πn/n) ), where 0<K<L. These (N/) complex numbers are multpled by the elements of the nput mage or by prevous complex numbers durng the computaton of the D-FFT. To create a complex number requres two real floatng pont operatons. So, the total number of computaton steps requred for the fast neural networks becomes: σ=(q+1)(5n log N ) +6qN +q(n - )+qn +N (9) whch can be reformulated as: σ=(q+1)(5n log N ) +q(8n -n ) +N (10) 6. Usng a sldng wndow of sze nxn for the same mage of NxN pxels, q(n -1)(N-n+1) computaton steps are requred when usng tradtonal neural networks for object detecton process. The theoretcal speed up factor η can be evaluated as follows: q(n -1)(N- n+ 1) η = (11) (q+ 1)(5N log N ) q(8n - n + ) + N The theoretcal speed up rato Eq. 11 wth dfferent szes of the nput mage and dfferent n sze weght matrces s lsted n Table 1. Practcal speed up rato for manpulatng mages of dfferent szes and dfferent n sze weght matrces s lsted n Table usng.7 GHz processor and MATLAB. An nterestng property wth FNNs s that the number of computaton steps does not depend on ether the sze of the nput sub-mage or the sze of the weght matrx (n). The effect of (n) on the number of computaton steps s very small and can be gnored. Ths s n contrast to CNNs n whch the number of computaton steps s ncreased wth the sze of both the nput sub-mage and the weght matrx (n). 3. Image Compresson by usng K-PCA By mplementng PCA t s known that, we should buld one partcular transformaton matrx consstng of egenvectors. So, when reconstructng the mage, not only the transformed coeffcents but also the transform matrx s requred. Furthermore PCA s a lnear approach; t cannot approxmate all areas of the mage equally well. In other words, one PCA cannot smultaneously capture the features of all regons. To resolve the above problems, MPC has been studed [- 7]. The procedure s as follows: before PCA, dvde the problem space nto a number of sub-spaces, and then fnd a set of egenvectors for each sub-space. If enough tranng data are gven, MPC can construct a system whch mantans a good generalty. It s nterestng to note that an MPC can be used as a unversal encoder f the generalzaton ablty s hgh enough. In ths case, we do not have to preserve the MPC parameters n the compressed data. Only the transformed coeffcents (the output of the system for each nput mage block) are needed. So far researches have been focused on how to dvde the problem space effcently. In [7,8], Donny proposed an optmally adaptve transform codng method. It s composed of a number of GHA neural networks. Fg. 1 ISSN: ISBN:

4 llustrates how the approprate GHA s selected to learn from the current nput vector. The tranng algorthm s as follows [7,8]: Step 1: Intalze (at random) K transformaton matrcesw, W, 1 L, W K, where W j s the weght matrx of the j-th GHA network. Step : For each tranng nput vector x, classfy t to the -th sub-space, f P = W T where. W Px K max P x = (1) j=1 Update the weghts accordng to the followng rule: W new old old = W + αz( x, W ) (13) Where α s the learnng rate and Z s a GHA learnng rule whch converges to the prncpal components. Step 3: Iteratvely mplement the above tranng procedure untl the weghts are stable. In [7], the tranng parameters are: 1) the number of sub-spaces s 64 and ) the number of tranng teratons s 80,000. Note that to use the MPC as a unversal encoder; we must tran t usng many data. The above algorthm clearly s not good enough because t s too tme consumng. In [8], several methods were proposed to speed up the tranng process and decrease the dstorton. These methods nclude growth by class nserton, growth by components addton and tree structured network. The essenta l ssue s that the convergent speed of GHA s very slow [9-1]. To enhance the operaton of PCA, the authors [9-1] proposed K-PCA. The encodng and decodng procedure of the K-PCA method s gven n Fg.. Step 1: Dvde the nput mage nto n n small blocks (n=8). For the entre nput data, fnd an 8-D PCA encoder By so dong, the dmenson of the problem space can be reduced from 64 to 8. j Step : Fnd a codebook wth k (k=64 n our experments) code words usng the LBG algorthm, for the 8-D vectors obtaned n the last step, and record the ndex of each nput vector. Step 3: Based on the codebook, the problem space can be dvded nto k clusters. For each cluster, we can fnd an M-D (M=4) PCA encoder. Step 4: For each nput vector, compress t to an 8- D vector usng the PCA encoder found n Step 1, then fnd the ndex of the nearest code word found n Step, and fnally compress t to an M-D vector. The M-D vector along wth the ndex of the nearest code word s used as the code of the nput vector. The purpose of Step 1 s to reduce the computatonal cost of VQ. Through experments t has been found that an 8-D PCA encoder can represent the orgnal mage very well. The codebook obtaned based on the 8-D vectors performs almost the same as that obtaned from the orgnal 64-D vectors. The above encodng method s called k-pca. Note that f k-pca s traned by usng enough data, t can be used as a unversal encoder, and do not have to nclude the egenvectors nto the compressed data. Thus, the compresson rato can be ncreased. The reconstructon (decodng) procedure s as follows [9-1]: Step 1: Read n the codes one by one. Step : Fnd the bass vectors for the cluster specfed by the ndex, and transform the M-D vector back to the 8-D vector. Step 3: Transform the 8-D vector back to vector, and put t to the mage n order. 4. Conclusons n n -D A new fast algorthm for covert survellance has been presented. Ths has been acheved by performng cross correlaton n the frequency doman between nput mage and the nput weghts of fast neural networks (FNNs). It has been proved mathematcally and practcally that the number of computaton steps requred for the presented FNNs s less than that needed by conventonal neural networks (CNNs). For effcent mage compresson a hybrd approach called k-pca has been used. It s well traned unversal egenvectors act as a common transformaton matrx lke cosne functon n DCT, and the VQ has been used to dvde ISSN: ISBN:

5 the tranng data nto k clusters. A pre-pca has also been used to reduce the tme for buldng the VQ codebook. Smulaton results usng MATLAB has confrmed the theoretcal computatons. References [1] Y. Lnde, A. Buzo and R. M. Gray, "An Algorthm for Vector Quantzaton," IEEE Trans. On Communcatons, Vol. 8, No.1, pp.84-95, [] R. Klette, and Zamperon, "Handbook of mage processng operators," John Wley & Sonsltd, [3] E. Oja, "A smplfed neuron model as a prncpal component analyzer", J. Math. Bology 15, pp , 198. [4] S. Carrato, Neural networks for mage compresson, Neural Networks: Adv. and Appl. ed., Gelenbe Pub,North-Holland, Amsterdam, 199, pp [5] T. D. Sanger, "Optmal unsupervsed learnng n a sngle-layer lnear feedforward neural network", Neural Networks, pp , [6] S. Y. Kung and K. I. Damantaras, "A neural network learnng algorthm for adaptve prncpal component extracton (APEX)", n Proc. IEEE Int. Conf. Acoustcs, Speech, and Sgnal Processng 90, pp , (Al-burqurque, NM), Aprl [7] R. D. Dony, "Adaptve Transform Codng of Images Usng a Mxture of Prncpal Components". PhD thess, McMaster Unversty, Hamlton, Ontaro, Canada, July [8] R. D. Dony, "A Comparson of Hebban Learnng Methods for Image Compresson usng the Mxture of Prncpal Components Network" Proceedngs of SPIE, v 3307, Applcatons of Artfcal Neural Networks n Image Processng III, pp , [9] C. F. Lv, Q. F. Zhao, and Z. W. Lu, "A flexble non-lnear PCA encoder for stll mage compresson," Proc. 7th IEEE Internatonal Conference on Computer and Informaton Technology (CIT07), pp , Oct [10] C. F. Lv and Q. F. Zhao, "k-pca: a sem-unversal encoder for mage compresson," Internatonal Journal of Pervasve Computng and Communcatons, Vol. 3, No., pp. 05-0, 007. [11] C. F. Lv and Q. F. Zhao, "e-lbg: an extended LBG algorthm for constructng unversal k-pca," WSEA Trans. on Sgnal Processng, Issue 3, Vol. 1, pp , 005. [1] C. F. Lv and Q. Zhao, " A unversal PCA for Image Compresson," Proc. of the nternatonal Conf. on Embedded and Ubqutous, Nagasak, Japan, 005, pp [13] J. W. Cooley and J. W. Tukey, An algorthm for the machne calculaton of complex Fourer seres, Math. Comput. 19, 1965, pp [14] H. M. El-Bakry, and Q. Zhao, Speedng-up Normalzed Neural Networks for Face/Object Detecton, Machne Graphcs & Vson Journal (MG&V), vol. 14, No.1, 005, pp [15] H. M. El-Bakry, "A Novel Hgh Speed Neural Model for Fast Pattern Recognton," Soft Computng Journal, vol. 14, no. 6, 010, pp [16] H. M. El-Bakry, "A New Neural Desgn for Faster Pattern Detecton Usng Cross Correlaton and Matrx Decomposton," Neural World journal, Neural World Journal, 009, vol. 19, no., pp [17] H. M. El-Bakry, "An Effcent Algorthm for Pattern Detecton usng Combned Classfers and Data Fuson," Informaton Fuson Journal, vol. 11, 010, pp [18] H. M. El-Bakry, "New Faster Normalzed Neural Networks for Sub-Matrx Detecton usng Cross Correlaton n the Frequency Doman and Matrx Decomposton," Appled Soft Computng journal, vol. 8, ssue, March 008, pp [19] R. Feraud, O. Berner, J. E. Vallet, and M. Collobert, " A Fast and Accurate Face Detector for Indexaton of Face Images," Fourth IEEE Internatonal Conference on Automatc Face and Gesture Recognton, Grenoble, France, 8-30 March, 000. [0] H. Schnederman and T. Kanade, "Probablstc modelng of local appearance and spatal relatonshps for object recognton, " In IEEE Conference on Computer Vson and Pattern Recognton (CVPR), pp , SantaBarbara, CA, [1] H. A. Rowley, S. Baluja, and T. Kanade, " Neural Network - Based Face Detecton, " IEEE Trans. on Pattern Analyss and Machne Intellgence, Vol. 0, No. 1, pp. 3-38, ISSN: ISBN:

6 Fg. 1: Basc structure of the MPC. 8D PCA Input X 8D VQ Encode k-pca PCs k-pca Decode 8D VQ Output X Fg. : The flow-chat of the K-PCA [9-1]. Table 1: The theoretcal speed up rato for mages wth dfferent szes. ISSN: ISBN:

7 Image sze Speed up rato (n=0) Speed up rato (n=5) Speed up rato (n=30) 100x x x x x x x x x x x x x x x x x x x x Table : Practcal speed up rato for mages wth dfferent szes usng MATLAB. Image sze Speed up rato (n=0) Speed up rato (n=5) Speed up rato (n=30) 100x x x x x x x x x x x x x x x x x x x x ISSN: ISBN:

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

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

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

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

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

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

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k)

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k) ISSN 1749-3889 (prnt), 1749-3897 (onlne) Internatonal Journal of Nonlnear Scence Vol.17(2014) No.2,pp.188-192 Modfed Block Jacob-Davdson Method for Solvng Large Sparse Egenproblems Hongy Mao, College of

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

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

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

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

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

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

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 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

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

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

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

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

Comparison of Wiener Filter solution by SVD with decompositions QR and QLP

Comparison of Wiener Filter solution by SVD with decompositions QR and QLP Proceedngs of the 6th WSEAS Int Conf on Artfcal Intellgence, Knowledge Engneerng and Data Bases, Corfu Island, Greece, February 6-9, 007 7 Comparson of Wener Flter soluton by SVD wth decompostons QR and

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

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

MATH 567: Mathematical Techniques in Data Science Lab 8

MATH 567: Mathematical Techniques in Data Science Lab 8 1/14 MATH 567: Mathematcal Technques n Data Scence Lab 8 Domnque Gullot Departments of Mathematcal Scences Unversty of Delaware Aprl 11, 2017 Recall We have: a (2) 1 = f(w (1) 11 x 1 + W (1) 12 x 2 + W

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

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

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

Exercises. 18 Algorithms

Exercises. 18 Algorithms 18 Algorthms Exercses 0.1. In each of the followng stuatons, ndcate whether f = O(g), or f = Ω(g), or both (n whch case f = Θ(g)). f(n) g(n) (a) n 100 n 200 (b) n 1/2 n 2/3 (c) 100n + log n n + (log n)

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

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

COMPUTATIONALLY EFFICIENT WAVELET AFFINE INVARIANT FUNCTIONS FOR SHAPE RECOGNITION. Erdem Bala, Dept. of Electrical and Computer Engineering,

COMPUTATIONALLY EFFICIENT WAVELET AFFINE INVARIANT FUNCTIONS FOR SHAPE RECOGNITION. Erdem Bala, Dept. of Electrical and Computer Engineering, COMPUTATIONALLY EFFICIENT WAVELET AFFINE INVARIANT FUNCTIONS FOR SHAPE RECOGNITION Erdem Bala, Dept. of Electrcal and Computer Engneerng, Unversty of Delaware, 40 Evans Hall, Newar, DE, 976 A. Ens Cetn,

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

CS 3710: Visual Recognition Classification and Detection. Adriana Kovashka Department of Computer Science January 13, 2015

CS 3710: Visual Recognition Classification and Detection. Adriana Kovashka Department of Computer Science January 13, 2015 CS 3710: Vsual Recognton Classfcaton and Detecton Adrana Kovashka Department of Computer Scence January 13, 2015 Plan for Today Vsual recognton bascs part 2: Classfcaton and detecton Adrana s research

More information

On the Repeating Group Finding Problem

On the Repeating Group Finding Problem The 9th Workshop on Combnatoral Mathematcs and Computaton Theory On the Repeatng Group Fndng Problem Bo-Ren Kung, Wen-Hsen Chen, R.C.T Lee Graduate Insttute of Informaton Technology and Management Takmng

More information

BACKGROUND SUBTRACTION WITH EIGEN BACKGROUND METHODS USING MATLAB

BACKGROUND SUBTRACTION WITH EIGEN BACKGROUND METHODS USING MATLAB BACKGROUND SUBTRACTION WITH EIGEN BACKGROUND METHODS USING MATLAB 1 Ilmyat Sar 2 Nola Marna 1 Pusat Stud Komputas Matematka, Unverstas Gunadarma e-mal: lmyat@staff.gunadarma.ac.d 2 Pusat Stud Komputas

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

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

A NEW DISCRETE WAVELET TRANSFORM

A NEW DISCRETE WAVELET TRANSFORM A NEW DISCRETE WAVELET TRANSFORM ALEXANDRU ISAR, DORINA ISAR Keywords: Dscrete wavelet, Best energy concentraton, Low SNR sgnals The Dscrete Wavelet Transform (DWT) has two parameters: the mother of wavelets

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

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

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

Hiding data in images by simple LSB substitution

Hiding data in images by simple LSB substitution Pattern Recognton 37 (004) 469 474 www.elsever.com/locate/patcog Hdng data n mages by smple LSB substtuton Ch-Kwong Chan, L.M. Cheng Department of Computer Engneerng and Informaton Technology, Cty Unversty

More information

CHALMERS, GÖTEBORGS UNIVERSITET. SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS. COURSE CODES: FFR 135, FIM 720 GU, PhD

CHALMERS, GÖTEBORGS UNIVERSITET. SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS. COURSE CODES: FFR 135, FIM 720 GU, PhD CHALMERS, GÖTEBORGS UNIVERSITET SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS COURSE CODES: FFR 35, FIM 72 GU, PhD Tme: Place: Teachers: Allowed materal: Not allowed: January 2, 28, at 8 3 2 3 SB

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

Novel Pre-Compression Rate-Distortion Optimization Algorithm for JPEG 2000

Novel Pre-Compression Rate-Distortion Optimization Algorithm for JPEG 2000 Novel Pre-Compresson Rate-Dstorton Optmzaton Algorthm for JPEG 2000 Yu-We Chang, Hung-Ch Fang, Chung-Jr Lan, and Lang-Gee Chen DSP/IC Desgn Laboratory, Graduate Insttute of Electroncs Engneerng Natonal

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

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

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

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

CSE4210 Architecture and Hardware for DSP

CSE4210 Architecture and Hardware for DSP 4210 Archtecture and Hardware for DSP Lecture 1 Introducton & Number systems Admnstratve Stuff 4210 Archtecture and Hardware for DSP Text: VLSI Dgtal Sgnal Processng Systems: Desgn and Implementaton. K.

More information

THE SUMMATION NOTATION Ʃ

THE SUMMATION NOTATION Ʃ Sngle Subscrpt otaton THE SUMMATIO OTATIO Ʃ Most of the calculatons we perform n statstcs are repettve operatons on lsts of numbers. For example, we compute the sum of a set of numbers, or the sum of the

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

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

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

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

Scroll Generation with Inductorless Chua s Circuit and Wien Bridge Oscillator

Scroll Generation with Inductorless Chua s Circuit and Wien Bridge Oscillator Latest Trends on Crcuts, Systems and Sgnals Scroll Generaton wth Inductorless Chua s Crcut and Wen Brdge Oscllator Watcharn Jantanate, Peter A. Chayasena, and Sarawut Sutorn * Abstract An nductorless Chua

More information

NUMERICAL DIFFERENTIATION

NUMERICAL DIFFERENTIATION NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the

More information

Tensor Subspace Analysis

Tensor Subspace Analysis Tensor Subspace Analyss Xaofe He 1 Deng Ca Partha Nyog 1 1 Department of Computer Scence, Unversty of Chcago {xaofe, nyog}@cs.uchcago.edu Department of Computer Scence, Unversty of Illnos at Urbana-Champagn

More information

On the correction of the h-index for career length

On the correction of the h-index for career length 1 On the correcton of the h-ndex for career length by L. Egghe Unverstet Hasselt (UHasselt), Campus Depenbeek, Agoralaan, B-3590 Depenbeek, Belgum 1 and Unverstet Antwerpen (UA), IBW, Stadscampus, Venusstraat

More information

Fourier Transform. Additive noise. Fourier Tansform. I = S + N. Noise doesn t depend on signal. We ll consider:

Fourier Transform. Additive noise. Fourier Tansform. I = S + N. Noise doesn t depend on signal. We ll consider: Flterng Announcements HW2 wll be posted later today Constructng a mosac by warpng mages. CSE252A Lecture 10a Flterng Exampel: Smoothng by Averagng Kernel: (From Bll Freeman) m=2 I Kernel sze s m+1 by m+1

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

Structure and Drive Paul A. Jensen Copyright July 20, 2003

Structure and Drive Paul A. Jensen Copyright July 20, 2003 Structure and Drve Paul A. Jensen Copyrght July 20, 2003 A system s made up of several operatons wth flow passng between them. The structure of the system descrbes the flow paths from nputs to outputs.

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

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

1 GSW Iterative Techniques for y = Ax

1 GSW Iterative Techniques for y = Ax 1 for y = A I m gong to cheat here. here are a lot of teratve technques that can be used to solve the general case of a set of smultaneous equatons (wrtten n the matr form as y = A), but ths chapter sn

More information

Pattern Matching Based on a Generalized Transform [Final Report]

Pattern Matching Based on a Generalized Transform [Final Report] Pattern Matchng ased on a Generalzed Transform [Fnal Report] Ram Rajagopal Natonal Instruments 5 N. Mopac Expwy., uldng, Austn, T 78759-354 ram.rajagopal@n.com Abstract In a two-dmensonal pattern matchng

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

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

FFT Based Spectrum Analysis of Three Phase Signals in Park (d-q) Plane

FFT Based Spectrum Analysis of Three Phase Signals in Park (d-q) Plane Proceedngs of the 00 Internatonal Conference on Industral Engneerng and Operatons Management Dhaka, Bangladesh, January 9 0, 00 FFT Based Spectrum Analyss of Three Phase Sgnals n Park (d-q) Plane Anuradha

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

Natural Images, Gaussian Mixtures and Dead Leaves Supplementary Material

Natural Images, Gaussian Mixtures and Dead Leaves Supplementary Material Natural Images, Gaussan Mxtures and Dead Leaves Supplementary Materal Danel Zoran Interdscplnary Center for Neural Computaton Hebrew Unversty of Jerusalem Israel http://www.cs.huj.ac.l/ danez Yar Wess

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 neural network with localized receptive fields for visual pattern classification

A neural network with localized receptive fields for visual pattern classification Unversty of Wollongong Research Onlne Faculty of Informatcs - Papers (Archve) Faculty of Engneerng and Informaton Scences 2005 A neural network wth localzed receptve felds for vsual pattern classfcaton

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

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

Image Processing for Bubble Detection in Microfluidics

Image Processing for Bubble Detection in Microfluidics Image Processng for Bubble Detecton n Mcrofludcs Introducton Chen Fang Mechancal Engneerng Department Stanford Unverst Startng from recentl ears, mcrofludcs devces have been wdel used to buld the bomedcal

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

A Fast Fractal Image Compression Algorithm Using Predefined Values for Contrast Scaling

A Fast Fractal Image Compression Algorithm Using Predefined Values for Contrast Scaling Proceedngs of the World Congress on Engneerng and Computer Scence 007 WCECS 007, October 4-6, 007, San Francsco, USA A Fast Fractal Image Compresson Algorthm Usng Predefned Values for Contrast Scalng H.

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

Lossy Compression. Compromise accuracy of reconstruction for increased compression.

Lossy Compression. Compromise accuracy of reconstruction for increased compression. Lossy Compresson Compromse accuracy of reconstructon for ncreased compresson. The reconstructon s usually vsbly ndstngushable from the orgnal mage. Typcally, one can get up to 0:1 compresson wth almost

More information

An efficient algorithm for multivariate Maclaurin Newton transformation

An efficient algorithm for multivariate Maclaurin Newton transformation Annales UMCS Informatca AI VIII, 2 2008) 5 14 DOI: 10.2478/v10065-008-0020-6 An effcent algorthm for multvarate Maclaurn Newton transformaton Joanna Kapusta Insttute of Mathematcs and Computer Scence,

More information

Digital Signal Processing

Digital Signal Processing Dgtal Sgnal Processng Dscrete-tme System Analyss Manar Mohasen Offce: F8 Emal: manar.subh@ut.ac.r School of IT Engneerng Revew of Precedent Class Contnuous Sgnal The value of the sgnal s avalable over

More information

10-701/ Machine Learning, Fall 2005 Homework 3

10-701/ Machine Learning, Fall 2005 Homework 3 10-701/15-781 Machne Learnng, Fall 2005 Homework 3 Out: 10/20/05 Due: begnnng of the class 11/01/05 Instructons Contact questons-10701@autonlaborg for queston Problem 1 Regresson and Cross-valdaton [40

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

APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM IN FAULT DIAGNOSIS

APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM IN FAULT DIAGNOSIS Journal of Theoretcal and Appled Informaton Technology 005-01 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM

More information

Transform Coding. Transform Coding Principle

Transform Coding. Transform Coding Principle Transform Codng Prncple of block-wse transform codng Propertes of orthonormal transforms Dscrete cosne transform (DCT) Bt allocaton for transform coeffcents Entropy codng of transform coeffcents Typcal

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

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

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

Difference Equations

Difference Equations Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1

More information

THE EFFECT OF TORSIONAL RIGIDITY BETWEEN ELEMENTS ON FREE VIBRATIONS OF A TELESCOPIC HYDRAULIC CYLINDER SUBJECTED TO EULER S LOAD

THE EFFECT OF TORSIONAL RIGIDITY BETWEEN ELEMENTS ON FREE VIBRATIONS OF A TELESCOPIC HYDRAULIC CYLINDER SUBJECTED TO EULER S LOAD Journal of Appled Mathematcs and Computatonal Mechancs 7, 6(3), 7- www.amcm.pcz.pl p-issn 99-9965 DOI:.75/jamcm.7.3. e-issn 353-588 THE EFFECT OF TORSIONAL RIGIDITY BETWEEN ELEMENTS ON FREE VIBRATIONS

More information

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 493 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces you have studed thus far n the text are real vector spaces because the scalars

More information

The Expectation-Maximization Algorithm

The Expectation-Maximization Algorithm The Expectaton-Maxmaton Algorthm Charles Elan elan@cs.ucsd.edu November 16, 2007 Ths chapter explans the EM algorthm at multple levels of generalty. Secton 1 gves the standard hgh-level verson of the algorthm.

More information

Improvement of Histogram Equalization for Minimum Mean Brightness Error

Improvement of Histogram Equalization for Minimum Mean Brightness Error Proceedngs of the 7 WSEAS Int. Conference on Crcuts, Systems, Sgnal and elecommuncatons, Gold Coast, Australa, January 7-9, 7 3 Improvement of Hstogram Equalzaton for Mnmum Mean Brghtness Error AAPOG PHAHUA*,

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

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

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

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