הטכניון - מכון טכנולוגי לישראל Technion - Israel Institute of Technology Технион - израильский технологический институт ألتخنيون - معهد تكنولوجي لإسرائيل

02360941 - מבוא לרשתות עצביות 02360941 - Introduction to Neural Networks 02360941 - Introduction to Neural Networks 02360941 - Introduction to Neural Networks

חורף 2014-2015Winter 2014-2015Зима 2014-2015شتاء 2014-2015

סילבוס Syllabus Силабус خطة الدورة

Neural networks as computational models. The classification problem. The neuron as a classifier. Rosenblatt's perceptron.
Rosenblatt's convergence theorem. Minsky and Pappert's counterexample.
Class separability. Cover's counting theorem. Separability of random variables. Generalization according to Cover.
Nonlinear separability according to Vapnik and Chervonenkis. The VC dimension. Generalization according to Vapnik and Chervonenkis.
Mapping multi-variable binary functions by neural networks. Existence and uniquness theorems.
Support Vector Classification (SVC) and Support Vector Regression (SVR).
Continuous function approximation by neural networks. Regularization.
Learning and information storage in neural networks. The back propagation method. Hebbian learning
Associative memories: Hopfield's and Kanerva's models. Storage capacities. Error correction.
The dynamics of discrete networks. Convergence and oscillations.
Some real-world examples.