Introducing the Separability Matrix for ECOC coding

Slides:



Advertisements
Similar presentations
Query Classification Using Asymmetrical Learning Zheng Zhu Birkbeck College, University of London.
Advertisements

ICONIP 2005 Improve Naïve Bayesian Classifier by Discriminative Training Kaizhu Huang, Zhangbing Zhou, Irwin King, Michael R. Lyu Oct
Combining Classification and Model Trees for Handling Ordinal Problems D. Anyfantis, M. Karagiannopoulos S. B. Kotsiantis, P. E. Pintelas Educational Software.
Multi Hand Pose Recognition System using Kinect Depth Sensor: Application to Medical Image Navigation Oscar Lopes, Miguel Pousa, Miguel Reyes, Sergio Escalera,
Transductive Reliability Estimation for Kernel Based Classifiers 1 Department of Computer Science, University of Ioannina, Greece 2 Faculty of Computer.
BoVDW: Bag-of-Visual-and-Depth- Words for Gesture Recognition All rights reserved HuBPA© Human Pose Recovery and Behavior Analysis Antonio Hernández-Vela.
Probability-based Dynamic Time Warping for Gesture Recognition on RGB-D data All rights reserved HuBPA© Human Pose Recovery and Behavior Analysis Group.
Learning on Probabilistic Labels Peng Peng, Raymond Chi-wing Wong, Philip S. Yu CSE, HKUST 1.
Arbitrary Bit Generation and Correction Technique for Encoding QC-LDPC Codes with Dual-Diagonal Parity Structure Chanho Yoon, Eunyoung Choi, Minho Cheong.
Second order cone programming approaches for handing missing and uncertain data P. K. Shivaswamy, C. Bhattacharyya and A. J. Smola Discussion led by Qi.
SLIQ: A Fast Scalable Classifier for Data Mining Manish Mehta, Rakesh Agrawal, Jorma Rissanen Presentation by: Vladan Radosavljevic.
Reducing Multiclass to Binary LING572 Fei Xia Week 9: 03/04/08.
Chapter 3 Coding Theory.
Finding Compact Structural Motifs Presented By: Xin Gao Authors: Jianbo Qian, Shuai Cheng Li, Dongbo Bu, Ming Li, and Jinbo Xu University of Waterloo,
ECOC for Text Classification Hybrids of EM & Co-Training (with Kamal Nigam) Learning to build a monolingual corpus from the web (with Rosie Jones) Effect.
Using Error-Correcting Codes For Text Classification Rayid Ghani Center for Automated Learning & Discovery, Carnegie Mellon University.
Combining Labeled and Unlabeled Data for Multiclass Text Categorization Rayid Ghani Accenture Technology Labs.
Efficient Text Categorization with a Large Number of Categories Rayid Ghani KDD Project Proposal.
GPGPU Volume Classification using SimpleOpenCL Oscar Amoros, Sergio Escalera, Anna Puig and Maria Salamó Computer Vision Center, Universitat Autònoma de.
Discriminative Naïve Bayesian Classifiers Kaizhu Huang Supervisors: Prof. Irwin King, Prof. Michael R. Lyu Markers: Prof. Lai Wan Chan, Prof. Kin Hong.
Using Error-Correcting Codes For Text Classification Rayid Ghani This presentation can be accessed at
Machine Learning as Applied to Intrusion Detection By Christine Fossaceca.
Generalized Communication System: Error Control Coding Occurs In Right Column. 6.
Ger man Aerospace Center Gothenburg, April, 2007 Coding Schemes for Crisscross Error Patterns Simon Plass, Gerd Richter, and A.J. Han Vinck.
Faculty of Computer Science © 2006 CMPUT 229 Special-Purpose Codes Binary, BCD, Hamming, Gray, EDC, ECC.
Selective Sampling on Probabilistic Labels Peng Peng, Raymond Chi-Wing Wong CSE, HKUST 1.
Using Error-Correcting Codes For Text Classification Rayid Ghani Center for Automated Learning & Discovery, Carnegie Mellon University.
ENN: Extended Nearest Neighbor Method for Pattern Recognition
Comparing the Parallel Automatic Composition of Inductive Applications with Stacking Methods Hidenao Abe & Takahira Yamaguchi Shizuoka University, JAPAN.
1 Secure Cooperative MIMO Communications Under Active Compromised Nodes Liang Hong, McKenzie McNeal III, Wei Chen College of Engineering, Technology, and.
Active Learning for Class Imbalance Problem
Super-Orthogonal Space- Time BPSK Trellis Code Design for 4 Transmit Antennas in Fast Fading Channels Asli Birol Yildiz Technical University,Istanbul,Turkey.
Presented by Tienwei Tsai July, 2005
B. Krishna Mohan and Shamsuddin Ladha
CODING/DECODING CONCEPTS AND BLOCK CODING. ERROR DETECTION CORRECTION Increase signal power Decrease signal power Reduce Diversity Retransmission Forward.
Exploiting Context Analysis for Combining Multiple Entity Resolution Systems -Ramu Bandaru Zhaoqi Chen Dmitri V.kalashnikov Sharad Mehrotra.
A Novel Local Patch Framework for Fixing Supervised Learning Models Yilei Wang 1, Bingzheng Wei 2, Jun Yan 2, Yang Hu 2, Zhi-Hong Deng 1, Zheng Chen 2.
Greedy is not Enough: An Efficient Batch Mode Active Learning Algorithm Chen, Yi-wen( 陳憶文 ) Graduate Institute of Computer Science & Information Engineering.
DIGITAL COMMUNICATIONS Linear Block Codes
Spatio-temporal Pattern Queries M. Hadjieleftheriou G. Kollios P. Bakalov V. J. Tsotras.
SemiBoost : Boosting for Semi-supervised Learning Pavan Kumar Mallapragada, Student Member, IEEE, Rong Jin, Member, IEEE, Anil K. Jain, Fellow, IEEE, and.
RSVM: Reduced Support Vector Machines Y.-J. Lee & O. L. Mangasarian First SIAM International Conference on Data Mining Chicago, April 6, 2001 University.
Optimal Dimensionality of Metric Space for kNN Classification Wei Zhang, Xiangyang Xue, Zichen Sun Yuefei Guo, and Hong Lu Dept. of Computer Science &
Online Multiple Kernel Classification Steven C.H. Hoi, Rong Jin, Peilin Zhao, Tianbao Yang Machine Learning (2013) Presented by Audrey Cheong Electrical.
A NOVEL METHOD FOR COLOR FACE RECOGNITION USING KNN CLASSIFIER
Computational Approaches for Biomarker Discovery SubbaLakshmiswetha Patchamatla.
On Utillizing LVQ3-Type Algorithms to Enhance Prototype Reduction Schemes Sang-Woon Kim and B. John Oommen* Myongji University, Carleton University*
Perfect and Related Codes
Fast Query-Optimized Kernel Machine Classification Via Incremental Approximate Nearest Support Vectors by Dennis DeCoste and Dominic Mazzoni International.
***Classification Model*** Hosam Al-Samarraie, PhD. CITM-USM.
… Algo 1 Algo 2 Algo 3 Algo N Meta-Learning Algo.
An Improved Algorithm for Decision-Tree-Based SVM Sindhu Kuchipudi INSTRUCTOR Dr.DONGCHUL KIM.
Musical Genre Categorization Using Support Vector Machines Shu Wang.
Efficient Text Categorization with a Large Number of Categories Rayid Ghani KDD Project Proposal.
Chapter 15: Classification of Time- Embedded EEG Using Short-Time Principal Component Analysis by Nguyen Duc Thang 5/2009.
Present by: Fang-Hui Chu Large Margin Gaussian Mixture Modeling for Phonetic Classification and Recognition Fei Sha*, Lawrence K. Saul University of Pennsylvania.
SemiBoost : Boosting for Semi-supervised Learning Pavan Kumar Mallapragada, Student Member, IEEE, Rong Jin, Member, IEEE, Anil K. Jain, Fellow, IEEE, and.
Mustafa Gokce Baydogan, George Runger and Eugene Tuv INFORMS Annual Meeting 2011, Charlotte A Bag-of-Features Framework for Time Series Classification.
High resolution product by SVM. L’Aquila experience and prospects for the validation site R. Anniballe DIET- Sapienza University of Rome.
Feature learning for multivariate time series classification Mustafa Gokce Baydogan * George Runger * Eugene Tuv † * Arizona State University † Intel Corporation.
BIOINFORMATION A new taxonomy-based protein fold recognition approach based on autocross-covariance transformation - - 王红刚 14S
Experience Report: System Log Analysis for Anomaly Detection
Glenn Fung, Murat Dundar, Bharat Rao and Jinbo Bi
Discriminative Training of Chow-Liu tree Multinet Classifiers
Combining Base Learners
Open-Category Classification by Adversarial Sample Generation
Information Redundancy Fault Tolerant Computing
Discriminative Frequent Pattern Analysis for Effective Classification
Paper ID: XX Track: Track Name
Concave Minimization for Support Vector Machine Classifiers
Presentation transcript:

Introducing the Separability Matrix for ECOC coding Miguel Angel Bautista, Sergio Escalera, Xavier Baró & Oriol Pujol

Outline Classification problems and the ECOC framework. Motivation. The Separability Matrix. The Confusion-Separability Extension Coding. Experiments and Results. Conclusions and Future Work.

Introduction to the ECOC framework In classification tasks, the goal is to classify and object among a certain number of possible categories. This framework is composed of two different steps : Coding : Decompose a given N-class problem into a set of n binary problems. Decoding : Given a test sample s, determine its category. + + + +

Introduction to the ECOC framework At the decoding step a new sample s is classified by comparing the binary responses to the rows of M by means of a decoding measure . Different types of decoding based on the distance used (i.e. Hamming, Euclidean, etc.)

Motivation Standard predefined strategies may not be suitable for a given problem. Literature often suggested the use of equidistant codes. In [1] we show how reduced codes can perform as well as standard designs with far less number of dichotomizers. Those reduced codes can be extended in a problem-dependent way to benefit from error-correcting principles. How can we identify those codes? [1] Compact Evolutionary Design of Error Correcting Output Codes, M. Bautista, S. Escalera, X. Baró, O. Pujol, J. Vitrià and P. Radeva

The Separability Matrix The Separability matrix S contains the pairwise distance between the codewords in M. With this matrix we can analyze the correction capability of M since, Standard ECOC designs shown constant Separability matrices. Identify those classes that need an increment of distance in order to benefit from error-correcting principles.

The CSE Coding Algorithm Find the most confused classes : Compute and Extension matrix E which increments Fill the empty Extension codes taking into account the confusion of with the rest of the classes. Update Confusion and Separability matrices.

Experiments: Data We tested the novel methodology on several public datasets from the UCI Machine Learning Repository. In addition we perform experiments with 3 public Computer Vision problems. The ARFace dataset with 20 classes. The Traffic Sign dataset with 36 classes. The MPEG dataset with 70 classes.

Experiments and Results Results for UCI and Computer Vision experiments with SVM as the base classifier.

Conclusions and Future Work The Separability Matrix is introduced as a novel tool to analyze and enhance ECOC coding designs. The Extension Algorithm proposed can be applied to any existing ECOC scheme. A new coding design based on the Separability matrix is introduced obtaining significant performance improvements over state-of-the-art ECOC designs. The proposed methodology reduces the number of base classifiers needed in comparison with state-of-the-art designs.

Thank you!