CS855 Overview Dr. Charles Tappert
Pattern Recognition and Machine Learning Immensely broad subject Applications in many fields Scene analysis, document searching, handwriting and gesture recognition, speech recognition and understanding, geological analysis, recognition of bubble chamber tracks, and biometrics Central to human-machine interface problems Siri – speech recognition and speech understanding
Pattern Classification by Duda, Hart, and Stork Definition: Pattern recognition is the act of taking in raw data and taking an action based on the “category” of the pattern Provides unified presentation of classification theory Parametric and nonparametric methods Supervised and unsupervised learning Discusses relative strengths and weaknesses of the various classification techniques
Deep Learning by Goodfellow, Bengio, and Courville Deep learning involves a hierarchy of concepts that allows the computer to learn complicated concepts by building them out of simpler ones Graphically the concepts are built on top of each other with many layers Deep learning solves the representation problem by introducing representations expressed in terms of simpler representations The quintessential example of a deep learning model is the feedforward deep network or MLP
Deep Learning: Practical Approach by Patterson and Gibson Deep learning systems are neural networks with a large number of parameters and layers in one of four fundamental architectures Unsupervised pretrained, convolutional, recurrent, and recursive networks First four chapters cover the theory and fundamentals Last five chapters cover a series of practical paths for building deep learning systems
This Course Each textbook has enough material for a two-semester course This course will cover Duda: most chapters – the important ones for the course focus in more depth than others Goodfellow: important chapters, with focus on the convolutional layers Patterson: important chapters, with focus on the practical aspects