Scribd
Upload
7 views

Deep Learning - Introduction To The Course

This document outlines a course on deep learning. It discusses the reasons for taking the course, including the availability of large datasets and job opportunities in AI. The educational objectives are to understand popular deep learning architectures like AlexNet, ResNet, Inception and GANs. The course will provide hands-on experience with frameworks like PyTorch, Keras and TensorFlow. The outline lists topics like convolutional neural networks, recurrent neural networks, autoencoders and GANs. There will be 12 lab sessions covering areas such as object detection, semantic segmentation and reinforcement learning. Students will be evaluated through assignments, quizzes, a midterm exam, end of semester exam and a semester project.

Uploaded by

Muhammad Waleed Khan
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views

Deep Learning - Introduction To The Course

This document outlines a course on deep learning. It discusses the reasons for taking the course, including the availability of large datasets and job opportunities in AI. The educational objectives are to understand popular deep learning architectures like AlexNet, ResNet, Inception and GANs. The course will provide hands-on experience with frameworks like PyTorch, Keras and TensorFlow. The outline lists topics like convolutional neural networks, recurrent neural networks, autoencoders and GANs. There will be 12 lab sessions covering areas such as object detection, semantic segmentation and reinforcement learning. Students will be evaluated through assignments, quizzes, a midterm exam, end of semester exam and a semester project.

Uploaded by

Muhammad Waleed Khan
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Deep Learning

Dr. Muhammad Tariq Saeed


<[email protected]>

School of Interdisciplinary Engineering and Sciences (SINES),


National University of Sciences and Technology(NUST)

1
Outline

• Why this course?


• Educational Objectives
• Learning Outcomes
• Course Outline
• Books
• Lab Sessions
• Grading

2
Why this Course?

• Availability of huge data sets


• Global AI boom
• High Performance Computing (HPC) as a drive force
– Progress in modern processor Architecture
– Availability of libraries, frameworks and tools
• Result
– Job Opportunities
– New Businesses

3
Educational Objectives

• The aim of this course is to understand deep


neural networks including multiple well-known
architectures (Alexnet, Resnet, Inception, and
GAN).

• Special emphasis on gaining hands-on


capability on latest deep learning tools (Pytorch,
Keras, Tensorflow, Colab). Vulnerabilities and
limitations of deep networks will be discussed
4
Course Outline
• Introduction to artificial neural networks
• How to train deep networks
• Introduction to Tensorflow I
• Introduction to Tensorflow I
• Introduction to Tensorflow II
• Project 1
• Introduction to CNN (AlexNet, Resnet)
• Introduction to Pytorch I
• Introduction to Pytorch II
• Processing sequences using RNNs
• Autoencoders & GANs
• Adversarial attacks on Deep Networks
• Adversarial attacks on Deep Networks II
• Project 2
• Reinforcement learning , Federated Learning
• Discussion on semester project/ research papers 5
Lab Sessions
• Lab Session 1: Introduction to Neural Networks
• Lab Session 2: Convolution Neural Networks
• Lab Session 3: Recurrent Neural Networks (RNNs)
• Lab Session 4: Transfer Learning
• Lab Session 5: Generative Adversarial Networks (GANs)
• Lab Session 6: Autoencoders
• Lab Session 7: Attention Mechanisms
• Lab Session 8: Sequence-to-Sequence Models
• Lab Session 9: Object Detection
• Lab Session 10: Semantic Segmentation
• Lab Session 11: Anomaly Detection
• Lab Session 12: Reinforcement Learning

6
Data Driven Learning

7
Traditional Approach

8
Machine Learning Approach

9
Machine Learning / Deep Learning Tasks

Clustering

https://www.javatpoint.com/regression-vs-classification-in-machine-learning

10
Schedule of Classes / Labs

Sr. Session Timings


1 Lecture 9:00 – 10:00 hrs.
2 Lab 11:00 – 12:50 hrs.

11
Books

• Deep Learning, Ian Goodfellow, Yoshua Bengio, Aaron


Courville, Francis Bach
• References/Resources will be shared along with course
material

12
Distribution of Marks / Grading

Sr. Activity Qty. Weightage/ Remarks


Percentage
1 Mid-Term 1 25% Paper-based Evaluation

2 End Semester Exam (ESE) 1 25% + 15%


(Project) Hybrid Evaluation

3 Assignments [Lab based] – 5 10% Speed


Group [2-3 students] based Programming/Computer-
Submission based Evaluation along
with Viva
4 Quizzes [Lab based] – 3 10% Computer based
Submission on LMS/Teams Evaluation
5 Semester Project 1 15% Computer based
Submission on LMS/Teams Evaluation with Final
Presentation

Paper-based Computer based


50% 50%
13
Contact Hours

• Wednesday (15:00 – 16:00 hrs.)


• Queries [email protected]

14

You might also like