1. Introduction Mohammad Beigi Department of Biomedical Engineering Isfahan University Majid.beigi@eng.ui.ac.ir

2. 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 ?

3. 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

4. Midterm 25% Final 40% Computer assignments 10% Final Programming Project 15% Seminar 10% Evaluation

5. 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

6. 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.

7. 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.

9. Pattern Recognition Applications

17. Figure 9: Clustering of Microarray Data

18. Pattern Recognition Applications Figure 10: Brain Control Interface

35. Regression: Polynomial Curve Fitting is continuous

36. Sum-of-Squares Error Function Optimization Problem

37. 0 th Order Polynomial

38. 1 st Order Polynomial

39. 3 rd Order Polynomial

40. 9 th Order Polynomial

41. Over-fitting Root-Mean-Square (RMS) Error:

42. Polynomial Coefficients

43. Data Set Size: 9 th Order Polynomial

44. Data Set Size: 9 th Order Polynomial

45. Regularization ;ridge regression Penalize large coefficient values Shrinkage: reduce the order of method

46. Regularization:

48. Regularization: vs.

49. Polynomial Coefficients Optimization Problem: Finding optimum

50. Classification example: Handwritten Digit Recognition 28*28 Pixel image  : 784 real numbers, training set:

54. Pattern recognition approaches

55. Statistical Pattern recognition

57. Structural Pattern Recognition

58. Neural Pattern Recognition