Deep Neural Networks UT, Fall 2023

AI
Course Overview

Deep Learning is a subset of Machine Learning and this field has some amazing real-world applications such as Apple’s Siri to Google’s Self Driving Car. Deep learning allows machines to solve relatively complex problems even when using data that is diverse, less structured or interdependent. Deep learning is a form of machine learning that is inspired and modeled on how the human brain works.

In this course students are introduced to the architecture of deep neural networks, algorithms that are developed to extract high-level feature representations of data. In addition to theoretical foundations of neural networks, including backpropagation and stochastic gradient descent, students get hands-on experience building deep neural network models with Python and learn how to use application program interfaces (APIs), such as TensorFlow and Keras, for building a variety of deep neural networks: convolutional neural network (CNN), recurrent neural network (RNN), self-organizing maps (SOM), generative adversarial network (GANs), and long short-term memory (LSTM).

Instructor

Class Time and Location
Time Sunday, Tuesday 9:00am -10:30 am
Location Class 206, Faculty of Mechanics
Grading
Midterm 20 %
Final 20 %
Homeworks&Project 60 %
Course Channel
Course Outline
Week Date Lecture Topic Slides Homework/Projects
1 9/18 Introduction Session_1
2 9/25
4 10/9 Fully Connected Neural Networks Session_2 HW_1
7 11/7 Convolutional Neural Networks Session_3
8 11/12 HW_Extra
9 11/14 HW_2
10 11/22 Region based CNNs Session_4
11 11/27 HW_3
12 12/1 Recurrent Neural Networks Session_5
13 12/9 HW_4
14 12/21 Transformers Session_6 HW_5
15 1/4 Variational Auto encoders and Generative Adversarial Networks Session_7 HW_6