Welcome to AIL7021 - Deep Learning



Instructor

TAs

Sudipan Saha

Kausik Hira, Poonam Rajput, Ritik Kumar Das, Palaash Goel


Course Logistics

Lectures : Monday and Thursday 2:00-3:30 pm

Different resources about the course are posted on Moodle.

Course Components and Grading Policy

  • Minor (20%), Major (35%)
  • Programming Assignments (30%)
  • Quiz (two) (15%)

Audit Policy
  • Equivalent to B-


Lecture-wise Topics

  • Lecture 1 - Introduction and M-P Neuron
  • Lecture 2 - Perceptron (including learning algorithm)
  • Lecture 3 - Perceptron convergence theorem, XOR example, deeper networks
  • Lecture 4 - Gradient descent and backpropagation
  • Lecture 5 - Backpropagation (continued) and loss functions
  • Lecture 6 - Regularization and dropout
  • Lecture 7 - Dropout (contd.) and batch normalization
  • Lecture 8 - Convolution basics
  • Lecture 9 - Convolution (continued), transposed convolution
  • Lecture 10 - Some CNN architectures
  • Lecture 11 - ResNet
  • Lecture 12 - MobileNet, Weight Initialization
  • Lecture 13 - Optimization: Momentum, AdaGrad, Adam
  • Lecture 14 - Mixed precision, Grad-CAM
  • Lecture 15 - Semantic segmentation (U-Net), Spatial Attention
  • Lecture 16 - GCN
  • Lecture 17 - NLP: N-gram model, subword tokenization
  • Lecture 18 - WaveNet, RNN, BPTT
  • Lecture 19 - LSTM, Machine Translation
  • Lecture 20 - GAN
  • Lecture 21 - Conditional GAN, CycleGAN
  • Lecture 22 - Diffusion
  • Lecture 23 - Diffusion (contd.), Transformer
  • Lecture 24 - Transformer (contd.). Knowledge distillation
  • Lecture 25 - Self-supervised learning
  • Lecture 26 - BERT, CLIP

Announcements

Please check Moodle.