Lectures

You can download the lectures here. We will try to upload lectures prior to their corresponding classes.

• 

  Introduction and Logistics
  tl;dr: Short introduction to the course and its logistics
  [notes] [slides]
  

  Suggested Readings:

  • Few things to know about Machine Learning
  • How Alpha Go works

  • Useful Stable diffusion prompts

  • Deep learning frameworks
• 

  Python and Numpy refresher.
  tl;dr: Setting up the conda env and refresher on Python
  [notes]
  

  Suggested Readings:

  • How to setup up Conda for deep learning
  • How to work with google Colab
  • Managing Conda environement
• 

  Machine Learning Refresher
  tl;dr: Image Classificationa and Data Driven algorithms
  [notes] [students_notes]
  

  Suggested Readings:

  • MNIST dataset
  • Loss functions visualizatio
  • Visualize optimization algorithms
  • CIFAR-10 dataset.
• 

  Image Classification problem
  tl;dr: The problem of Image classification
  [notes] [slides]
  

  Suggested Readings:

  • MNIST dataset
  • Fast Library for Approxmiation Nearest neighbors
  • Recognizing and Learning Object Categories Deep Learning using Linear Support Vector Machines
• 

  Optimisation Algorithms
  tl;dr: Optimisation algorithms
  [notes]
  

  Suggested Readings:

  • Why Momentum works
  • Adaptive SubGradient methods
  • Adam: A method for stochatic optimization
  • On the convergence of the Adam method
• 

  Mulit-layers Perceptrons
  tl;dr: Multi layers Perceptrons
  [notes]
  

  Suggested Readings:

  • Tensor MLP playground
  • Activations function in neural networks
  • Dropout
• 

  Pytorch Review Session
  tl;dr: Pytorch
  [notes]
  

  Suggested Readings:

  • Pytorch tutorials
  • Deep Learning with PyTorch: A 60 Minute Blitz
  • Pytorch Custom Model Step by Step
  • Create a custom dataloader
  • Building a custom data loader
• 

  Convolutional neural networks.
  tl;dr: Convolution
  [notes]
  

  Suggested Readings:

  • Soumith benchmarks for CONV performance
  • ConvNetJS CIFAR-10 demo allows you to play with ConvNet architectures and see the results and computations in real time, in the browser.
  • Caffe, one of the popular ConvNet libraries.
  • State of the art ResNets in Torch7
• 

  Modern Convolutional Networks
  tl;dr: Modern Conv
  [notes]
  

  Suggested Readings:

  • Multi Object Tracking using Transformers