Artificial Neural Networks

Artificial Neural Networks

Sat. and Mon. from 08:00 am to 09:30 am

 Room #3, Abourayhan Building

2^nd Semester 1397-1398 (Jan. 2019)

Instructor: Dr. Mohammad Reza Ahmadzadeh

Room Location: ECE  Building - 322
Internal Telephone: Ext. 5370
External Telephone: Iran: +98 (0)31  33915370
Fax: Iran: +98 (0)31 33912451
Email:

Ahmadzadeh @ iut.ac.ir

Ahmadzadeh.m @ gmail.com

Office Hours:

I will try to be in my office on Saturdays, Mondays and Wednesdays from 9 :30 to 12 :00 am and but I will always do my best to be available for students by appointment or any time that I am free.

Course Description:

This course covers the fundamentals of Artificial Neural Networks.

Outline:

•Introduction

–Linear algebra review

–Probability and calculus chain rules

–Introduction to machine learning 

•Multilayer Perceptrons

•Logistic regression and MLP

•Backpropagation 

•Stochastic gradient descent

•Optimization 

Generalization

•Radial-basis function networks

Convolutional Networks

•Recurrent Neural Nets 

Textbook:

•Charu C. Aggarwal, Neural Networks and Deep Learning. A Textbook-(2018)

•Martin T. Hagan, Howard B. Demuth, Mark Beale, Orlando De Jesús, Neural Network Design. 2014.

•Simon Haykin, Neural Networks and Learning Machines, 3rd Edition, 2009.

•I. Goodfellow, Y. Bengio, A. Courville, Deep Learning 2016.

•Christopher M. Bishop, Neural Networks for Pattern Recognition, 1995.

•Michael Nielsen, Neural Networks and Deep Learning, 2017 (Online).

•Ke-Lin Du, M. N. S. Swamy, Neural Networks and Statistical Learning, Springer London 2014.

Optional:

Lecture Notes:

Slides modified by Instructor and lecture notes can be obtained via E-Learning LMS (Enrolled students, Password protected).

Link: http://lms.iut.ac.ir/

Prerequisites: 

Linear algebra: basic matrix operations, span, rank, range and null space, eigenvalue decomposition, singular value decomposition, pseudoinverse

Probability: independence, conditional probability, Bayes rule, multivariate Gaussian distribution, marginalization, expectation, variance

Grading Policy: 

Midterm Exam ≈ 25%

HW, Comp. Assignments and projects: ≈ 30%

Final exam ≈ 45%

Prerequisites: 

Linear algebra: basic matrix operations, span, rank, range and null space, eigenvalue decomposition, singular value decomposition, pseudoinverse

Probability: independence, conditional probability, Bayes rule, multivariate Gaussian distribution, marginalization, expectation, variance

Grading Policy: 

Midterm Exam ≈ 25%

HW, Comp. Assignments and projects: ≈ 30%

Final exam ≈ 45%

Time: 

8:00 - 9:30 Sat. Mon.

Term: 

2019

Grade: 

Graduate