Foundations of Neural Networks and Deep Learning: Basics of image/signal like convolution – Perceptron – Multi Layer Perceptron – Feedforward Neural Networks – Back propagation, Training and Testing Neural Networks, Activation Functions, Loss Functions, Hyperparameters. Optimization algorithms and Regularization – Batch Normalization.

Deep Neural Networks: Common Architectural Principles of Deep Networks, Building Blocks of Deep Networks.

Major Architectures of Deep Networks: Convolutional Neural Networks (CNNs), Deep CNN, Recurrent Neural Networks, Recursive Neural Networks, Adversarial learning, Transfer Learning, Generative AI.

Linear factor models, Autoencoders, Representation learning, Structured probabilistic models, Monte-Carlo models. Tensorflow/Pytorch: DenoisingSparsity in Autoencoders Models for Sequence Analysis –Vanishing Gradients Long Short-Term Memory (LSTM) Units- Primitives for RNN Models– Named Entity Recognition/Opinion Mining/Sentiment Analysis/Question Answering/Neural Summarization. Seminar : Cutting edge technologies like LLMs.

Pre-Requisite(s): None
Course Type: NLP

Course Objectives

• To introduce the basics of Neural networks and Deep neural networks.
• To explore the major deep network architectures including convolutional and recurrent.
• To enable and apply different deep neural network models.
• To apply various deep learning techniques to provide solutions for real world problems.

Course Outcomes

CO1: Understand the foundational concepts behind Neural Networks, Deep Learning and Deep Neural architectures.

CO2: Apply different frameworks to develop deep neural network architecture.

CO3: Apply different pre-trained and custom models to analyze the performance.

CO4: Design and deploy appropriate deep learning solutions for real-world problems.

CO-PO Mapping

Evaluation Pattern: 70/30

Note: Continuous assessments can include quizzes, tutorials, lab assessments, case study and project reviews. Midterm and End semester exams can be a theory exam or lab integrated exam for two hours

1. Ian Goodfellow, YoshuaBengio and Aeron Courville, Deep Learning, MIT Press,First Edition, 2016.
2. Francois Chollet, Deep Learning with Python, Manning Publications Co, First Edition, 2018.
3. Gibson and Josh Patterson, Deep Learning A practitioner’s approach, Adam O’Reilly, First Edition, 2017.
4. Nikhil Buduma, “Fundamentals of Deep Learning: Designing Next-Generation Machine Intelligence Algorithm”, O’Reilly, 2017
5. Nikhil Ketkar, “Deep Learning with Python: A Hands-on Introduction”, Apress, 2017.