Introduction Mohammad Beigi Department of Biomedical Engineering Isfahan University
Pattern recognition and Machine Learning Syllabus Introduction, Linear Models for classification Neural Networks (MLP, RBF, SOM, LVQ, ADALINE) Kernel Methods & Support Vector Machines Statistical Pattern Recognition ? (HMM,EM, Clustering and unsupervised learning ? Feature Selection and Dimension reduction ?
Pattern recognition and Machine Learning Texts R. O. Duda, P. E. Hart, D. G. Stork, Pattern Classification, 2nd edition, John Wiley & Sons, Inc., 2000.Pattern Classification M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.Pattern Recognition and Machine Learning
Midterm 25% Final 40% Computer assignments 10% Final Programming Project 15% Seminar 10% Evaluation
Human Perception Humans have developed highly sophisticated skills for sensing their environment and taking actions according to what they observe, e.g Understanding spoken words reading handwriting distinguishing fresh food from its smell We would like to give similar capabilities to machines
What is Pattern Recognition? A pattern is an entity, vaguely defined, that could be given a name, e.g., fingerprint image, handwritten word, human face, speech signal, DNA sequence, Pattern recognition is the study of how machines can observe the environment, learn to distinguish patterns of interest, make sound and reasonable decisions about the categories of the patterns.
Human and Machine Perception We are often influenced by the knowledge of how patterns are modeled and recognized in nature when we develop pattern recognition algorithms. Research on machine perception also helps us gain deeper understanding and appreciation for pattern recognition systems in nature. Yet, we also apply many techniques that are purely numerical and do not have any correspondence in natural systems.
Pattern Recognition Applications
Figure 9: Clustering of Microarray Data
Pattern Recognition Applications Figure 10: Brain Control Interface
Regression: Polynomial Curve Fitting is continuous
Sum-of-Squares Error Function Optimization Problem
0 th Order Polynomial
1 st Order Polynomial
3 rd Order Polynomial
9 th Order Polynomial
Over-fitting Root-Mean-Square (RMS) Error:
Polynomial Coefficients
Data Set Size: 9 th Order Polynomial
Data Set Size: 9 th Order Polynomial
Regularization ;ridge regression Penalize large coefficient values Shrinkage: reduce the order of method
Regularization:
Regularization: vs.
Polynomial Coefficients Optimization Problem: Finding optimum
Classification example: Handwritten Digit Recognition 28*28 Pixel image : 784 real numbers, training set:
Pattern recognition approaches
Statistical Pattern recognition
Structural Pattern Recognition
Neural Pattern Recognition