Course Information
Course Code
21CS71
Credits
03
Total Hours
40 hours
Examination
Theory (3 hours)
Course Objectives
- Understand the fundamentals of deep learning.
- Know the theory behind Convolutional Neural Networks, Autoencoders, RNN.
- Illustrate the strengths and weaknesses of many popular deep learning approaches.
- Introduce major deep learning algorithms, the problem settings, and their applications to solve real-world problems.
- Learn the open issues in deep learning, and grasp the current research directions.
Course Modules
Module 1: Introduction to Deep Learning
Topics: Introduction, Deep learning Model, Historical Trends in Deep Learning,Learning Algorithms, Supervised Learning Algorithms, Unsupervised Learning Algorithms.
Module 2: Feedforward Networks
Topics: Introduction to feedforward neural networks, Gradient-Based Learning, Back- Propagation and Other Differentiation Algorithms.
Module 3: Optimization for Training Deep Models:
Topics: Empirical Risk Minimization, Challenges in Neural Network Optimization, Basic Algorithms: Stochastic Gradient Descent, Parameter Initialization Strategies,
Module 4: Convolutional Networks
Topics: The Convolution Operation, Pooling, Convolution and Pooling as an Infinitely Strong Prior, Variants of the Basic Convolution Function, Structured Outputs, Data Types, Efficient Convolution Algorithms, Random or Unsupervised Features- LeNet, AlexNet
Module 5: Recurrent and Recursive Neural Networks
Topics: Unfolding Computational Graphs, Recurrent Neural Network, Bidirectional RNNs, Deep Recurrent Networks, Recursive Neural Networks, The Long Short- Term Memory and Other Gated RNNs.