CCS417 – Artificial Intelligence: Deep Learning

📅 Schedule 📝 Assignments 📊 Grading Policy 📚 Resources 👥 Staff 🎓 Students

This course provides a comprehensive introduction to modern deep learning architectures and neural network fundamentals. Designed for 4th-year computer science undergraduates, students will explore foundational neural models, advanced optimization techniques, generative modeling, and attention-based architectures. Practical implementation and theoretical understanding are combined over an intensive 8-week curriculum. Prerequisites: Linear algebra, basic probability, and Python programming.

Schedule

Type	Group	Time and Venue
Lecture	Group A	Mondays 1:00 - 2:00 pm, Senate Hall.
Lecture	Group B	Mondays 2:30 - 3:30 pm, Senate Hall.
Lab	Group A	Tuesdays 2:00 - 3:00 pm, Main Computer Lab.
Lab	Group B	Tuesdays 3:00 - 4:00 pm, Main Computer Lab.
Lab	Group C	Thursdays 2:00 - 3:00 pm, Main Computer Lab.
Lab	Group D	Thursdays 3:00 - 4:00 pm, Main Computer Lab.

Lecture

Week	Module Topic	Description
Week 1 (09/14/2026)	Foundations of Deep Learning (slides)	Introduction to AI, Deep Learning, and Course Logistics.
Week 2 (09/21/2026)	Artificial Neural Networks (Slides)	Introduction to neural networks, perceptrons, activation functions, and feedforward dynamics.
Week 3 (09/28/2026)	Optimization and Training (Slides)	Gradient descent, backpropagation algorithms, momentum, Adam optimizer, and learning rate scheduling. Overfitting prevention, dropout, weight decay (L1/L2), batch normalization, and data augmentation..
Week 4 (10/12/2026)	Convolutional Neural Networks (Slides)	Spatial convolutions, pooling layers, feature maps, and modern CNN architectures (ResNet, VGG).
Week 5 (10/19/2026)	RNN & LSTM Networks (Slides)	Sequential data modeling, recurrent neural networks, vanishing gradients, and Long Short-Term Memory units.
Week 6 (10/26/2026)	Transformers and Attention	Self-attention mechanisms, multi-head attention, positional encoding, and transformer architectures.
Week 7 (11/02/2026)	Generative Models	Autoencoders, Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and diffusion models.
Week 8 (11/09/2026)	Advanced Topics	Reinforcement learning integration, model deployment, efficient inference, and state-of-the-art frontiers.

Labs

Week	Lab Topic	Description
Week 2 (09/21/2026)	Creating Artificial Neural Networks Click here launch to Colab	Learn how to create Artificial Neural Networks (ANNs) in Google Colab using Tensorflow/Keras.
Week 3 (09/28/2026)	Training and Optimizing Neural Networks Click here to launch Colab	Learn how to train and optimize ANNs (forward and backward propagation)
Week 4 (10/05/2026)	Creating Convolutional Neural Networks Click here to launch Colab	Learn how to create a CNN model to extract patterns from images.
Week 5 (10/12/2026)	Training and Evaluating a CNN Click here to luanch Colab	Learn how to train and evaluate a CNN on Fashion data
Week 6 (10/19/2026)	Transfer Learning	Learn how to leverage knowledge of pre-trained models to solve custom problems/tasks.
Week 7 (10/26/2026)	Performance Metrics	Learn how to compute metrics to evaluate the performance of a model.
Week 8 (11/02/2026)	Contest	Compete in group of 3-5 to build a model for a given task (task to be unveiled).

Assignments

Assignment 1: Class Registration
Course registration and environment setup setup task. – Click here to complete registration

Assignment 2: Lecture 1 assessment
Due September 15, 2026 - Quiz to assess your comprehension of lecture 1 materials.

Assignment 3: Lecture 2 assessment
Due September 22, 2026 - Quiz to assess your comprehension of lecture 2 materials.

Assignment 4: Lecture 3 assessment
Due September 29, 2026 - Quiz to assess your comprehension of lecture 3 materials.

Assignment 5: Lecture 4 assessment
Due October 6, 2026 - Quiz to assess your comprehension of lecture 4 materials.

Project Deliverable 1: Expressions – Click here to submit your expressions

Grading Policy

10%

Attendance

20%

Assignments

70%

Final Exam

📌 All assignments must be submitted on time. Late submissions incur a 10% penalty per day. The final exam includes written theoretical assessment and practical coding evaluations.