Generalized Transfer Component Analysis for Mismatched Jpeg Steganalysis
|
|
- Cory Hart
- 5 years ago
- Views:
Transcription
1 Generalized Transfer Component Analysis for Mismatched Jpeg Steganalysis Xiaofeng Li 1, Xiangwei Kong 1, Bo Wang 1, Yanqing Guo 1,Xingang You 2 1 School of Information and Communication Engineering Dalian University of Technology, Dalian, , China 2 Beijing Institute of Electronic Technology and Application Beijing, , China 1
2 Contents 1 Motivation of Our Work 2 Proposed Method 3 Experiments and Results 4 Summary and Future Work 2
3 Contents 1 Motivation of Our Work 2 Proposed Method 3 Experiments and Results 4 Summary and Future Work 3
4 Motivation of Our Work Battle of steganography and steganalysis Steganography embed message signal into cover images to get stego images; message undetectable in covert communication steganalysis features sensitive to change due to embedding build decision model using machine learning recognize stego images from plain cover images Steganalysis seem to win the battle recently [Kodovsky and Fridrich 12]: Rich model perform well to detect six modern steganographic schemes at low embedding rate. 4
5 Motivation of Our Work Steganalysis Really Win the Battle? Success of state of the art steganalysis methods rely on having prior knowledge of steganography to build the training set. cover images, embedding algorithm is known. Matched steganalysis train set and test set: matched cover images matched embedding algorithm laboratory Mismatched steganalysis train set and test set: mismatched cover images mismatched embedding algorithm real world 5
6 Motivation of Our Work Steganalysis Really Win the battle? [1] [Kodovsky and Fridrich 12]:Steganalysis of jpeg images using rich models [2] [Tomas Pevny and Jessica Fridrich 07]:Merging markovand dct features for multi-class jpeg steganalysis 6
7 Motivation of Our Work New Challenge in Steganalysis State-of-the-art steganalysis method could not be used effectively in the real world. Moving steganalysis from the laboratory to the real world. [Andrew D. Ker, Patrick Bas, Rainer Bohme, Remi Cogranne, Scott Craver Tomas Filler, Jessica Fridrich, Tomas Pevny 13]: Moving steganography And steganalysis from the laboratory to the real world. laboratory matched How? Real world mismatched 7
8 Related Work Motivation of Our Work [Ivans Lubenko, Andrew D. Ker 13]: Steganalysis with Mismatched Covers: Do Simple Classifiers Help? Large data large samples diverse enough rich features Simple classifier Ensemble Fisher Linear Discriminant Online Ensemble Average Perceptron general model Limitation: It costs much labor to collect images for such a training set. Can we train a model robust to mismatched steganalysis using a small set of samples? only a single set, not diverse, small number. 8
9 Contents 1 Motivation of Our Work 2 Proposed Method 3 Experiments and Results 4 Summary and Future Work 9
10 Main idea Proposed Method Train set: Source domain D = {( x, y ), i = 1, 2,, N} ~ P ( X, Y ) S i i S Test set: Target domain D = {( x,?), i = 1, 2,, M} ~ P ( X, Y ) T i T The two distributions are not the same! The two distributions are similar! 10
11 Proposed Method domain adaptation & transfer learning For Mismatch in other area: Natural Language Processing Object recognition Ex. [J. Blitzer et al EMNLP 2006] Ex. [R. Gopalan et al ICCV 2011] Video analysis Text classification Ex. [Jeff Donahue et al CVPR 2013] Ex. [Pan et al IEEE Tran-NN 2011] domain adaptation & transfer learning Learning a shared representation Assumption: a latent feature space exists in which classification hypotheses fit both domains. min P ( f ( X ), Y )- P ( f ( X ), Y ) f S T 11
12 Proposed Method Main idea Such a latent feature space leads to loss of some information, and may not be sensitive to embedding. According to target domain, transform source domain to an intermediate domain. Then find a latent feature space between target domain and intermediate domain. 12
13 Proposed Method Generalized Transfer Component Analysis Domain Alignment: transform source domain to intermediate domain. Learn Shared Feature Space: find a latent feature space between target and intermediate domain. Map Samples into the Feature Space Construct Classifier and Make Decision in the New Feature Space 13
14 Domain Alignment Proposed Method The aim is to transform source to an intermediate that is close to target. similar to 0-1 normalization liner transformation to hold the feature sensitivity to different categories. Objective : E ( ϕ ( X ), Y ) = E ( X, Y ) s σ ( ϕ ( X ), Y ) = σ ( X, Y ) s i i σ ( X t, yi ) ϕ ( xs ) = ( xs E ( X s, yi )) + E ( X t, yi ) σ ( X, y ) No labels in test set (target domain). We can t get the values. s t t i 14
15 Domain Alignment Proposed Method Objective : Liner transformation : E ( ψ ( X )) = E( X ) s s t σ ( ψ ( X )) = σ ( X ) s i i σ ( X t ) ϕ( xs ) = ( xs E( X s )) + E( X t ) σ ( X ) s t Train Model p ( y x ) s t t 1 i i E( Xt Y) xt p( y xt ) i p( y x ) σ 1 t i 2 i ( Xt Y) ( xt - E( Xt, Y)) p( y xt ) i p( y xt ) 15
16 Proposed Method Find shared feature space Objective: Simplify: min P ( f ( X ), Y )- P ( f ( X ), Y ) f P( X, Y ) = P( Y X ) P( X ) P ( Y X ) = P ( Y X ) s S t T min P ( f ( X))- P ( f ( X)) f S T Measure the distance of two distribution: ns nt 1 i 1 i Dis( PS ( X ), PT ( X )) = φ( xs ) φ( xt ) ns i= 1 nt i= 1 Dis( P ( X ), P ( X )) = trace( KL) S T φ (.) RHKS K. M. Borgwardt, A. Gretton, M. J. Rasch, H. P. Kriegel, B. Scholkopf, and A. J. Smola, Integrating structured biological data by kernel maximum mean discrepancy, Bioinformatics,
17 Proposed Method Find shared feature space define a non-liner kernel feature extraction matrix W as transformation: X = KW Update the new K: Update the new distance: T Dis( P ( X ), P ( X )) = trace( KL) = trace( KWW KL) S T new T T K new = X new X new = KWW K min trace( KWW T KL) W S. Pan, I. Tsang, J. Kwok, and Q.Yang, Domain adap-tation via transfer component analysis, IEEE Trans-actions on Neural Networks,
18 Proposed Method Find shared feature space To avoid the solution W=0, we add a constrain that which can preserve (or maximize) the initial data variance in the new space: T W KHKW The final kernel learning problem is then set up as: T T min tr( W W ) + µ tr( KWW KL) W = T s. t. W KHKW = I 1 W ( I µ KLK ) KHK I + (M leading eigenvectors) S. Pan, I. Tsang, J. Kwok, and Q.Yang, Domain adap-tation via transfer component analysis, IEEE Trans-actions on Neural Networks,
19 Proposed Method source domain Domain alignment intermediate domain target domain w Mapped source Mapped target Train model 19
20 Contents 1 Motivation of Our Work 2 Proposed Method 3 Experiments and Results 4 Summary and Future Work
21 Experiments and Results Experimental Setup Database: eight mismatched domains F5 Jsteg MBS A-F A-J A-M Set A: 1800 Outguess F5 A-O B-F Each domain Cover:450 Stego:450 Jsteg B-J MBS B-M Set B:1800 Outguess B-O 21
22 Experiments and Results Experimental Setup Database: eight mismatched domains Features: PF274 + our approach (GTCA) Classifier: lib-svm Approach compared with: Orig-Fea: PF-274+ lib-svm [Pevny and Fridrich 07]:Merging markov and dct features for multi-class jpeg steganalysis OEAP: JRM features + OEAP [Kodovsky and Fridrich 12]:Steganalysis of jpeg images using rich models [Ivans Lubenko, Andrew D. Ker 13]: Steganalysis with Mismatched Covers: Do Simple Classifiers Help? TCA: PF274+ TCA+ lib-svm [Pan et al 2011] Domain adap-tation via transfer component analysis 22
23 Experiments and Results Mismatched Experiment 1 Mismatched covers: different quantization table 23
24 Experiments and Results Mismatched Experiment 2 Mismatched stegos: different embedding algorithm 24
25 Experiments and Results Mismatched Experiment 3 Mismatched covers and stegos: different quantization table and different embedding algorithm 25
26 Contents 1 Motivation of Our Work 2 Proposed Method 3 Experiments and Results 4 Summary
27 Summary Mismatched steganalysis Important in real application Traditional steganalysis methods perform badly Two distributions are not the same Generalized Transfer Component Analysis (GTCA) Learn new representations to correct mismatches A small set of training samples Empirically successful New Strategy for Mismatched Steganalysis Domain adaptation, transfer learning 27
28 28
Information Hiding and Covert Communication
Information Hiding and Covert Communication Andrew Ker adk @ comlab.ox.ac.uk Royal Society University Research Fellow Oxford University Computing Laboratory Foundations of Security Analysis and Design
More informationComplete characterization of perfectly secure stego-systems with mutually independent embedding operation
Complete characterization of perfectly secure stego-systems with mutually independent embedding operation Tomáš Filler and Jessica Fridrich Dept. of Electrical and Computer Engineering SUNY Binghamton,
More informationTheoretical Model of the FLD Ensemble Classifier Based on Hypothesis Testing Theory
Theoretical Model of the FLD Ensemble Classifier Based on Hypothesis Testing Theory Rémi Cogranne, Member, IEEE, ICD - ROSAS - LM2S Troyes University of Technology - UMR 6281, CNRS 12, rue Marie Curie
More informationFisher Information Determines Capacity of ε-secure Steganography
Fisher Information Determines Capacity of ε-secure Steganography Tomáš Filler and Jessica Fridrich Dept. of Electrical and Computer Engineering SUNY Binghamton, New York 11th Information Hiding, Darmstadt,
More informationQuantitative Steganalysis of LSB Embedding in JPEG Domain
Quantitative Steganalysis of LSB Embedding in JPEG Domain Jan Kodovský, Jessica Fridrich September 10, 2010 / ACM MM&Sec 10 1 / 17 Motivation Least Significant Bit (LSB) embedding Simplicity, high embedding
More informationThe Square Root Law of Steganographic Capacity
The Square Root Law of Steganographic Capacity ABSTRACT Andrew D. Ker Oxford University Computing Laboratory Parks Road Oxford OX1 3QD, UK adk@comlab.ox.ac.uk Jan Kodovský Dept. of Electrical and Computer
More informationImperfect Stegosystems
DISSERTATION DEFENSE Imperfect Stegosystems Asymptotic Laws and Near-Optimal Practical Constructions Tomá² Filler Dept. of Electrical and Computer Engineering SUNY Binghamton, New York April 1, 2011 2
More informationSteganalysis of Spread Spectrum Data Hiding Exploiting Cover Memory
Steganalysis of Spread Spectrum Data Hiding Exploiting Cover Memory Kenneth Sullivan, Upamanyu Madhow, Shivkumar Chandrasekaran, and B.S. Manjunath Department of Electrical and Computer Engineering University
More informationOptimal Detection of OutGuess using an Accurate Model of DCT Coefficients
Optimal Detection of OutGuess using an Accurate Model of DCT Coefficients Thanh Hai Thai, Rémi Cogranne, Florent Retraint ICD - ROSAS - LMS - Troyes University of Technology - UMR 68, CNRS Troyes - France
More informationImproved Adaptive LSB Steganography based on Chaos and Genetic Algorithm
Improved Adaptive LSB Steganography based on Chaos and Genetic Algorithm Lifang Yu, Yao Zhao 1, Rongrong Ni, Ting Li Institute of Information Science, Beijing Jiaotong University, BJ 100044, China Abstract
More informationMultiple Similarities Based Kernel Subspace Learning for Image Classification
Multiple Similarities Based Kernel Subspace Learning for Image Classification Wang Yan, Qingshan Liu, Hanqing Lu, and Songde Ma National Laboratory of Pattern Recognition, Institute of Automation, Chinese
More informationLow Complexity Features for JPEG Steganalysis Using Undecimated DCT
1 Low Complexity Features for JPEG Steganalysis Using Undecimated DCT Vojtěch Holub and Jessica Fridrich, Member, IEEE Abstract This article introduces a novel feature set for steganalysis of JPEG images
More informationSTATISTICAL DETECTION OF INFORMATION HIDING BASED ON ADJACENT PIXELS DIFFERENCE
2th European Signal Processing Conference EUSIPCO 22) Bucharest, Romania, August 27-3, 22 STATISTICAL DETECTION OF INFORMATION HIDING BASED ON ADJACENT PIXELS DIFFERENCE Rémi Cogranne, Cathel Zitzmann,
More informationJPEG-Compatibility Steganalysis Using Block-Histogram of Recompression Artifacts
JPEG-Compatibility Steganalysis Using Block-Histogram of Recompression Artifacts Jan Kodovský and Jessica Fridrich Department of ECE, Binghamton University, NY, USA {fridrich,jan.kodovsky}@binghamton.edu
More informationDetection of Content Adaptive LSB Matching (a Game Theory Approach)
Detection of Content Adaptive LSB Matching (a Game Theory Approach) Tomáš Denemark, Jessica Fridrich / 5 Content-adaptive steganography Every pixel is changed with probability β i = exp( λρ i) + exp( λρ
More informationSteganalysis with a Computational Immune System
DIGITAL FORENSIC RESEARCH CONFERENCE Steganalysis with a Computational Immune System By Jacob Jackson, Gregg Gunsch, Roger Claypoole, Gary Lamont Presented At The Digital Forensic Research Conference DFRWS
More informationAdaptive Kernel Principal Component Analysis With Unsupervised Learning of Kernels
Adaptive Kernel Principal Component Analysis With Unsupervised Learning of Kernels Daoqiang Zhang Zhi-Hua Zhou National Laboratory for Novel Software Technology Nanjing University, Nanjing 2193, China
More informationCovariate Shift in Hilbert Space: A Solution via Surrogate Kernels
Kai Zhang kzhang@nec-labs.com NEC Laboratories America, Inc., 4 Independence Way, Suite 2, Princeton, NJ 854 Vincent W. Zheng Advanced Digital Sciences Center, 1 Fusionopolis Way, Singapore 138632 vincent.zheng@adsc.com.sg
More informationA Study of Embedding Operations and Locations for Steganography in H.264 Video
A Study of Embedding Operations and Locations for Steganography in H.264 Video Andreas Neufeld and Andrew D. Ker Oxford University Department of Computer Science Andreas Neufeld is now with the Image and
More informationUnsupervised Domain Adaptation with Distribution Matching Machines
Unsupervised Domain Adaptation with Distribution Matching Machines Yue Cao, Mingsheng Long, Jianmin Wang KLiss, MOE; NEL-BDS; TNList; School of Software, Tsinghua University, China caoyue1@gmail.com mingsheng@tsinghua.edu.cn
More informationResearch Article Improved Adaptive LSB Steganography Based on Chaos and Genetic Algorithm
Hindawi Publishing Corporation EURASIP Journal on Advances in Signal Processing Volume 2010, Article ID 876946, 6 pages doi:10.1155/2010/876946 Research Article Improved Adaptive LSB Steganography Based
More informationApprentissage, réseaux de neurones et modèles graphiques (RCP209) Neural Networks and Deep Learning
Apprentissage, réseaux de neurones et modèles graphiques (RCP209) Neural Networks and Deep Learning Nicolas Thome Prenom.Nom@cnam.fr http://cedric.cnam.fr/vertigo/cours/ml2/ Département Informatique Conservatoire
More informationDynamic Time-Alignment Kernel in Support Vector Machine
Dynamic Time-Alignment Kernel in Support Vector Machine Hiroshi Shimodaira School of Information Science, Japan Advanced Institute of Science and Technology sim@jaist.ac.jp Mitsuru Nakai School of Information
More information10/05/2016. Computational Methods for Data Analysis. Massimo Poesio SUPPORT VECTOR MACHINES. Support Vector Machines Linear classifiers
Computational Methods for Data Analysis Massimo Poesio SUPPORT VECTOR MACHINES Support Vector Machines Linear classifiers 1 Linear Classifiers denotes +1 denotes -1 w x + b>0 f(x,w,b) = sign(w x + b) How
More informationInstance-based Domain Adaptation via Multi-clustering Logistic Approximation
Instance-based Domain Adaptation via Multi-clustering Logistic Approximation FENG U, Nanjing University of Science and Technology JIANFEI YU, Singapore Management University RUI IA, Nanjing University
More informationMachine Learning for Structured Prediction
Machine Learning for Structured Prediction Grzegorz Chrupa la National Centre for Language Technology School of Computing Dublin City University NCLT Seminar Grzegorz Chrupa la (DCU) Machine Learning for
More informationKernel Methods. Barnabás Póczos
Kernel Methods Barnabás Póczos Outline Quick Introduction Feature space Perceptron in the feature space Kernels Mercer s theorem Finite domain Arbitrary domain Kernel families Constructing new kernels
More informationc 4, < y 2, 1 0, otherwise,
Fundamentals of Big Data Analytics Univ.-Prof. Dr. rer. nat. Rudolf Mathar Problem. Probability theory: The outcome of an experiment is described by three events A, B and C. The probabilities Pr(A) =,
More informationSemi-supervised Dictionary Learning Based on Hilbert-Schmidt Independence Criterion
Semi-supervised ictionary Learning Based on Hilbert-Schmidt Independence Criterion Mehrdad J. Gangeh 1, Safaa M.A. Bedawi 2, Ali Ghodsi 3, and Fakhri Karray 2 1 epartments of Medical Biophysics, and Radiation
More informationA Two-Stage Weighting Framework for Multi-Source Domain Adaptation
A Two-Stage Weighting Framework for Multi-Source Domain Adaptation Qian Sun, Rita Chattopadhyay, Sethuraman Panchanathan, Jieping Ye Computer Science and Engineering, Arizona State University, AZ 8587
More informationECE662: Pattern Recognition and Decision Making Processes: HW TWO
ECE662: Pattern Recognition and Decision Making Processes: HW TWO Purdue University Department of Electrical and Computer Engineering West Lafayette, INDIANA, USA Abstract. In this report experiments are
More informationMachine Learning : Support Vector Machines
Machine Learning Support Vector Machines 05/01/2014 Machine Learning : Support Vector Machines Linear Classifiers (recap) A building block for almost all a mapping, a partitioning of the input space into
More informationDomain Adaptation via Transfer Component Analysis
Domain Adaptation via Transfer Component Analysis Sinno Jialin Pan 1,IvorW.Tsang, James T. Kwok 1 and Qiang Yang 1 1 Department of Computer Science and Engineering Hong Kong University of Science and Technology,
More informationImproving Selection-Channel-Aware Steganalysis Features
Improving Selection-Channel-Aware Steganalysis Features Tomáš Denemark and Jessica Fridrich, Department of ECE, SUNY Binghamton, NY, USA, {tdenema1,fridrich}@binghamton.edu, Pedro Comesaña-Alfaro, Department
More informationCourse 10. Kernel methods. Classical and deep neural networks.
Course 10 Kernel methods. Classical and deep neural networks. Kernel methods in similarity-based learning Following (Ionescu, 2018) The Vector Space Model ò The representation of a set of objects as vectors
More informationQuantitative Structural Steganalysis of Jsteg
Quantitative Structural Steganalysis of Jsteg Jan Kodovský and Jessica Fridrich, Member, IEEE Abstract Quantitative steganalysis strives to estimate the change rate defined as the relative number of embedding
More informationCS4495/6495 Introduction to Computer Vision. 8C-L3 Support Vector Machines
CS4495/6495 Introduction to Computer Vision 8C-L3 Support Vector Machines Discriminative classifiers Discriminative classifiers find a division (surface) in feature space that separates the classes Several
More informationSupport Vector Machine. Industrial AI Lab.
Support Vector Machine Industrial AI Lab. Classification (Linear) Autonomously figure out which category (or class) an unknown item should be categorized into Number of categories / classes Binary: 2 different
More informationHoldout and Cross-Validation Methods Overfitting Avoidance
Holdout and Cross-Validation Methods Overfitting Avoidance Decision Trees Reduce error pruning Cost-complexity pruning Neural Networks Early stopping Adjusting Regularizers via Cross-Validation Nearest
More informationLogistic Regression. Machine Learning Fall 2018
Logistic Regression Machine Learning Fall 2018 1 Where are e? We have seen the folloing ideas Linear models Learning as loss minimization Bayesian learning criteria (MAP and MLE estimation) The Naïve Bayes
More informationDiscriminative Direction for Kernel Classifiers
Discriminative Direction for Kernel Classifiers Polina Golland Artificial Intelligence Lab Massachusetts Institute of Technology Cambridge, MA 02139 polina@ai.mit.edu Abstract In many scientific and engineering
More informationHYPERGRAPH BASED SEMI-SUPERVISED LEARNING ALGORITHMS APPLIED TO SPEECH RECOGNITION PROBLEM: A NOVEL APPROACH
HYPERGRAPH BASED SEMI-SUPERVISED LEARNING ALGORITHMS APPLIED TO SPEECH RECOGNITION PROBLEM: A NOVEL APPROACH Hoang Trang 1, Tran Hoang Loc 1 1 Ho Chi Minh City University of Technology-VNU HCM, Ho Chi
More informationAN INTRODUCTION TO NEURAL NETWORKS. Scott Kuindersma November 12, 2009
AN INTRODUCTION TO NEURAL NETWORKS Scott Kuindersma November 12, 2009 SUPERVISED LEARNING We are given some training data: We must learn a function If y is discrete, we call it classification If it is
More informationDistinguish between different types of scenes. Matching human perception Understanding the environment
Scene Recognition Adriana Kovashka UTCS, PhD student Problem Statement Distinguish between different types of scenes Applications Matching human perception Understanding the environment Indexing of images
More information6.036 midterm review. Wednesday, March 18, 15
6.036 midterm review 1 Topics covered supervised learning labels available unsupervised learning no labels available semi-supervised learning some labels available - what algorithms have you learned that
More informationNatural Language Processing. Classification. Features. Some Definitions. Classification. Feature Vectors. Classification I. Dan Klein UC Berkeley
Natural Language Processing Classification Classification I Dan Klein UC Berkeley Classification Automatically make a decision about inputs Example: document category Example: image of digit digit Example:
More informationUsing Both Latent and Supervised Shared Topics for Multitask Learning
Using Both Latent and Supervised Shared Topics for Multitask Learning Ayan Acharya, Aditya Rawal, Raymond J. Mooney, Eduardo R. Hruschka UT Austin, Dept. of ECE September 21, 2013 Problem Definition An
More informationImage Data Compression. Steganography and steganalysis Alexey Pak, PhD, Lehrstuhl für Interak;ve Echtzeitsysteme, Fakultät für Informa;k, KIT
Image Data Compression Steganography and steganalysis 1 Stenography vs watermarking Watermarking: impercep'bly altering a Work to embed a message about that Work Steganography: undetectably altering a
More informationImportance Reweighting Using Adversarial-Collaborative Training
Importance Reweighting Using Adversarial-Collaborative Training Yifan Wu yw4@andrew.cmu.edu Tianshu Ren tren@andrew.cmu.edu Lidan Mu lmu@andrew.cmu.edu Abstract We consider the problem of reweighting a
More informationClassification with Perceptrons. Reading:
Classification with Perceptrons Reading: Chapters 1-3 of Michael Nielsen's online book on neural networks covers the basics of perceptrons and multilayer neural networks We will cover material in Chapters
More informationUniversität Potsdam Institut für Informatik Lehrstuhl Maschinelles Lernen. Linear Classifiers. Blaine Nelson, Tobias Scheffer
Universität Potsdam Institut für Informatik Lehrstuhl Linear Classifiers Blaine Nelson, Tobias Scheffer Contents Classification Problem Bayesian Classifier Decision Linear Classifiers, MAP Models Logistic
More informationSupport Vector Machines. Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar
Data Mining Support Vector Machines Introduction to Data Mining, 2 nd Edition by Tan, Steinbach, Karpatne, Kumar 02/03/2018 Introduction to Data Mining 1 Support Vector Machines Find a linear hyperplane
More informationLinear Models for Classification
Linear Models for Classification Henrik I Christensen Robotics & Intelligent Machines @ GT Georgia Institute of Technology, Atlanta, GA 30332-0280 hic@cc.gatech.edu Henrik I Christensen (RIM@GT) Linear
More informationBrief Introduction to Machine Learning
Brief Introduction to Machine Learning Yuh-Jye Lee Lab of Data Science and Machine Intelligence Dept. of Applied Math. at NCTU August 29, 2016 1 / 49 1 Introduction 2 Binary Classification 3 Support Vector
More informationStatistical Pattern Recognition
Statistical Pattern Recognition Support Vector Machine (SVM) Hamid R. Rabiee Hadi Asheri, Jafar Muhammadi, Nima Pourdamghani Spring 2013 http://ce.sharif.edu/courses/91-92/2/ce725-1/ Agenda Introduction
More informationCluster Kernels for Semi-Supervised Learning
Cluster Kernels for Semi-Supervised Learning Olivier Chapelle, Jason Weston, Bernhard Scholkopf Max Planck Institute for Biological Cybernetics, 72076 Tiibingen, Germany {first. last} @tuebingen.mpg.de
More informationKernel methods for comparing distributions, measuring dependence
Kernel methods for comparing distributions, measuring dependence Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Principal component analysis Given a set of M centered observations
More informationLearning with kernels and SVM
Learning with kernels and SVM Šámalova chata, 23. května, 2006 Petra Kudová Outline Introduction Binary classification Learning with Kernels Support Vector Machines Demo Conclusion Learning from data find
More informationClassification Ensemble That Maximizes the Area Under Receiver Operating Characteristic Curve (AUC)
Classification Ensemble That Maximizes the Area Under Receiver Operating Characteristic Curve (AUC) Eunsik Park 1 and Y-c Ivan Chang 2 1 Chonnam National University, Gwangju, Korea 2 Academia Sinica, Taipei,
More informationFace Recognition Using Laplacianfaces He et al. (IEEE Trans PAMI, 2005) presented by Hassan A. Kingravi
Face Recognition Using Laplacianfaces He et al. (IEEE Trans PAMI, 2005) presented by Hassan A. Kingravi Overview Introduction Linear Methods for Dimensionality Reduction Nonlinear Methods and Manifold
More informationDetection of Content Adaptive LSB Matching (a Game Theory Approach)
Detection of Content Adaptive LSB Matching (a Game Theory Approach) Tomáš Denemark and Jessica Fridrich Department of ECE, SUNY Binghamton, NY, USA ABSTRACT This paper is an attempt to analyze the interaction
More informationAruna Bhat Research Scholar, Department of Electrical Engineering, IIT Delhi, India
International Journal of Scientific Research in Computer Science, Engineering and Information Technology 2017 IJSRCSEIT Volume 2 Issue 6 ISSN : 2456-3307 Robust Face Recognition System using Non Additive
More informationInformation Extraction from Text
Information Extraction from Text Jing Jiang Chapter 2 from Mining Text Data (2012) Presented by Andrew Landgraf, September 13, 2013 1 What is Information Extraction? Goal is to discover structured information
More informationSupport Vector Machines (SVM) in bioinformatics. Day 1: Introduction to SVM
1 Support Vector Machines (SVM) in bioinformatics Day 1: Introduction to SVM Jean-Philippe Vert Bioinformatics Center, Kyoto University, Japan Jean-Philippe.Vert@mines.org Human Genome Center, University
More informationDetecting LSB Matching by Applying Calibration Technique for Difference Image
Detecting LSB atching by Applying Calibration Technique for Difference Image iaolong Li Institute of Computer Science & Technology Peking University 87, Beijing, P. R. China lixiaolong@icst.pku.edu.cn
More informationIntroduction to Support Vector Machines
Introduction to Support Vector Machines Hsuan-Tien Lin Learning Systems Group, California Institute of Technology Talk in NTU EE/CS Speech Lab, November 16, 2005 H.-T. Lin (Learning Systems Group) Introduction
More informationSupport Vector Machines vs Multi-Layer. Perceptron in Particle Identication. DIFI, Universita di Genova (I) INFN Sezione di Genova (I) Cambridge (US)
Support Vector Machines vs Multi-Layer Perceptron in Particle Identication N.Barabino 1, M.Pallavicini 2, A.Petrolini 1;2, M.Pontil 3;1, A.Verri 4;3 1 DIFI, Universita di Genova (I) 2 INFN Sezione di Genova
More informationLeast Squares SVM Regression
Least Squares SVM Regression Consider changing SVM to LS SVM by making following modifications: min (w,e) ½ w 2 + ½C Σ e(i) 2 subject to d(i) (w T Φ( x(i))+ b) = e(i), i, and C>0. Note that e(i) is error
More informationSparse Support Vector Machines by Kernel Discriminant Analysis
Sparse Support Vector Machines by Kernel Discriminant Analysis Kazuki Iwamura and Shigeo Abe Kobe University - Graduate School of Engineering Kobe, Japan Abstract. We discuss sparse support vector machines
More informationInstance-based Domain Adaptation
Instance-based Domain Adaptation Rui Xia School of Computer Science and Engineering Nanjing University of Science and Technology 1 Problem Background Training data Test data Movie Domain Sentiment Classifier
More informationSupport Vector Machines using GMM Supervectors for Speaker Verification
1 Support Vector Machines using GMM Supervectors for Speaker Verification W. M. Campbell, D. E. Sturim, D. A. Reynolds MIT Lincoln Laboratory 244 Wood Street Lexington, MA 02420 Corresponding author e-mail:
More informationSupport Vector Machine. Industrial AI Lab. Prof. Seungchul Lee
Support Vector Machine Industrial AI Lab. Prof. Seungchul Lee Classification (Linear) Autonomously figure out which category (or class) an unknown item should be categorized into Number of categories /
More informationLinking non-binned spike train kernels to several existing spike train metrics
Linking non-binned spike train kernels to several existing spike train metrics Benjamin Schrauwen Jan Van Campenhout ELIS, Ghent University, Belgium Benjamin.Schrauwen@UGent.be Abstract. This work presents
More informationLearning the Semantic Correlation: An Alternative Way to Gain from Unlabeled Text
Learning the Semantic Correlation: An Alternative Way to Gain from Unlabeled Text Yi Zhang Machine Learning Department Carnegie Mellon University yizhang1@cs.cmu.edu Jeff Schneider The Robotics Institute
More informationSupport Vector Machines. CAP 5610: Machine Learning Instructor: Guo-Jun QI
Support Vector Machines CAP 5610: Machine Learning Instructor: Guo-Jun QI 1 Linear Classifier Naive Bayes Assume each attribute is drawn from Gaussian distribution with the same variance Generative model:
More informationBoosting: Algorithms and Applications
Boosting: Algorithms and Applications Lecture 11, ENGN 4522/6520, Statistical Pattern Recognition and Its Applications in Computer Vision ANU 2 nd Semester, 2008 Chunhua Shen, NICTA/RSISE Boosting Definition
More informationPattern Recognition and Machine Learning. Perceptrons and Support Vector machines
Pattern Recognition and Machine Learning James L. Crowley ENSIMAG 3 - MMIS Fall Semester 2016 Lessons 6 10 Jan 2017 Outline Perceptrons and Support Vector machines Notation... 2 Perceptrons... 3 History...3
More informationMinOver Revisited for Incremental Support-Vector-Classification
MinOver Revisited for Incremental Support-Vector-Classification Thomas Martinetz Institute for Neuro- and Bioinformatics University of Lübeck D-23538 Lübeck, Germany martinetz@informatik.uni-luebeck.de
More informationLearning with Noisy Labels. Kate Niehaus Reading group 11-Feb-2014
Learning with Noisy Labels Kate Niehaus Reading group 11-Feb-2014 Outline Motivations Generative model approach: Lawrence, N. & Scho lkopf, B. Estimating a Kernel Fisher Discriminant in the Presence of
More informationSGN (4 cr) Chapter 5
SGN-41006 (4 cr) Chapter 5 Linear Discriminant Analysis Jussi Tohka & Jari Niemi Department of Signal Processing Tampere University of Technology January 21, 2014 J. Tohka & J. Niemi (TUT-SGN) SGN-41006
More informationSupport Vector Machines
Support Vector Machines Le Song Machine Learning I CSE 6740, Fall 2013 Naïve Bayes classifier Still use Bayes decision rule for classification P y x = P x y P y P x But assume p x y = 1 is fully factorized
More informationObtaining Higher Rates for Steganographic Schemes while Maintaining the Same Detectability
Obtaining Higher Rates for Steganographic Schemes while Maintaining the Same Detectability Anindya Sarkar, Kaushal Solanki and and B. S. Manjunath Department of Electrical and Computer Engineering, University
More informationMULTIPLEKERNELLEARNING CSE902
MULTIPLEKERNELLEARNING CSE902 Multiple Kernel Learning -keywords Heterogeneous information fusion Feature selection Max-margin classification Multiple kernel learning MKL Convex optimization Kernel classification
More informationLecture 13 Visual recognition
Lecture 13 Visual recognition Announcements Silvio Savarese Lecture 13-20-Feb-14 Lecture 13 Visual recognition Object classification bag of words models Discriminative methods Generative methods Object
More informationMIRA, SVM, k-nn. Lirong Xia
MIRA, SVM, k-nn Lirong Xia Linear Classifiers (perceptrons) Inputs are feature values Each feature has a weight Sum is the activation activation w If the activation is: Positive: output +1 Negative, output
More informationLinear & nonlinear classifiers
Linear & nonlinear classifiers Machine Learning Hamid Beigy Sharif University of Technology Fall 1396 Hamid Beigy (Sharif University of Technology) Linear & nonlinear classifiers Fall 1396 1 / 44 Table
More informationDimension Reduction (PCA, ICA, CCA, FLD,
Dimension Reduction (PCA, ICA, CCA, FLD, Topic Models) Yi Zhang 10-701, Machine Learning, Spring 2011 April 6 th, 2011 Parts of the PCA slides are from previous 10-701 lectures 1 Outline Dimension reduction
More informationarxiv: v1 [stat.ml] 17 Dec 2018
Domain Adaptation on Graphs by Learning Graph Topologies: Theoretical Analysis and an Algorithm arxiv:1812.6944v1 [stat.ml] 17 Dec 218 Elif Vural 1 1 Department of Electrical and Electronics Engineering,
More informationLinear Classification. CSE 6363 Machine Learning Vassilis Athitsos Computer Science and Engineering Department University of Texas at Arlington
Linear Classification CSE 6363 Machine Learning Vassilis Athitsos Computer Science and Engineering Department University of Texas at Arlington 1 Example of Linear Classification Red points: patterns belonging
More informationNumerical Learning Algorithms
Numerical Learning Algorithms Example SVM for Separable Examples.......................... Example SVM for Nonseparable Examples....................... 4 Example Gaussian Kernel SVM...............................
More informationA GENERAL FORMULATION FOR SUPPORT VECTOR MACHINES. Wei Chu, S. Sathiya Keerthi, Chong Jin Ong
A GENERAL FORMULATION FOR SUPPORT VECTOR MACHINES Wei Chu, S. Sathiya Keerthi, Chong Jin Ong Control Division, Department of Mechanical Engineering, National University of Singapore 0 Kent Ridge Crescent,
More informationUnsupervised Domain Adaptation with Residual Transfer Networks
Unsupervised Domain Adaptation with Residual Transfer Networks Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I. Jordan KLiss, MOE; TNList; School of Software, Tsinghua University, China University
More informationA short introduction to supervised learning, with applications to cancer pathway analysis Dr. Christina Leslie
A short introduction to supervised learning, with applications to cancer pathway analysis Dr. Christina Leslie Computational Biology Program Memorial Sloan-Kettering Cancer Center http://cbio.mskcc.org/leslielab
More informationBrief Introduction of Machine Learning Techniques for Content Analysis
1 Brief Introduction of Machine Learning Techniques for Content Analysis Wei-Ta Chu 2008/11/20 Outline 2 Overview Gaussian Mixture Model (GMM) Hidden Markov Model (HMM) Support Vector Machine (SVM) Overview
More informationJeff Howbert Introduction to Machine Learning Winter
Classification / Regression Support Vector Machines Jeff Howbert Introduction to Machine Learning Winter 2012 1 Topics SVM classifiers for linearly separable classes SVM classifiers for non-linearly separable
More informationMachine learning for pervasive systems Classification in high-dimensional spaces
Machine learning for pervasive systems Classification in high-dimensional spaces Department of Communications and Networking Aalto University, School of Electrical Engineering stephan.sigg@aalto.fi Version
More informationMatrix Embedding with Pseudorandom Coefficient Selection and Error Correction for Robust and Secure Steganography
Matrix Embedding with Pseudorandom Coefficient Selection and Error Correction for Robust and Secure Steganography Anindya Sarkar, Student Member, IEEE, Upamanyu Madhow, Fellow, IEEE, and B. S. Manjunath,
More informationA Multi-task Learning Strategy for Unsupervised Clustering via Explicitly Separating the Commonality
A Multi-task Learning Strategy for Unsupervised Clustering via Explicitly Separating the Commonality Shu Kong, Donghui Wang Dept. of Computer Science and Technology, Zhejiang University, Hangzhou 317,
More informationWhat is semi-supervised learning?
What is semi-supervised learning? In many practical learning domains, there is a large supply of unlabeled data but limited labeled data, which can be expensive to generate text processing, video-indexing,
More informationDiscriminative Models
No.5 Discriminative Models Hui Jiang Department of Electrical Engineering and Computer Science Lassonde School of Engineering York University, Toronto, Canada Outline Generative vs. Discriminative models
More information