Lectures

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

• 

  Course Introduction
  tl;dr: An introduction to the course syllabus, timeline and background.
  [slides]
  
• 

  Concept Learning
  tl;dr: Learning from Examples, General-to-specific ordering over hypotheses, Version Spaces & candidate elimination algo, Picking new examples, the need for inductive bias
  [Slides] [Scribe]
  
• 

  Concept Learning: Continued
  tl;dr: Find-S Algorithm, Version Space and CANDIDATE- ELIMINATION, LIST-THEN-ELIMINATION Algorithm, Overview to Desicion Tree.
  [Slides]
  
• 

  Decision Tree
  tl;dr: Desicion Tree: Introduction, Useage, Types, Shanon's Entropy, Information Gain, Examples.
  [Slides]
  
• 

  Decision Tree: Continued
  tl;dr: Desicion Tree: ID3, Inductive Bias, Overfitting & it's avoiding techniques
  [Slides]
  
• 

  Linear Regression-I
  tl;dr: Linear Regression: MLE, MEP, Introduction to LR
  [Scribe]
  
• 

  Linear Regression-II
  tl;dr: Linear Regression: Normal equation, closed form solution for LR
  [proof] [scribe]
  

  Supplementary Reading:

  • Matrix Properties
• 

  Linear Regression-III
  tl;dr: Probabilistic interpretation of LR, Locally weighted LR, analysis of different error functions in LR
  [boardwork] [notes] [scribe]
  
• 

  Logistic Regression
  tl;dr: Logistic regression, training of logistic regression, sigmoid function, concave objective function
  [boardwork] [scribe] [reading-I] [reading-II]
  

  Supplementary Reading:

  • Bishop 4.0-4.5
• 

  Generalized Linear Models (GLMs)
  tl;dr: Exponential family of distributions, link function, Proofs of Normal/Poisson/Binomial distributions in GLMs
  [boardwork] [reading-I]
  

  Supplementary Reading:

  • Kavin Murphy (Chapter 12)
• 

  Generative Learning Algorithms
  tl;dr: Generative Learning algorithm, Gaussian discriminant analysis (Introduction)
  [boardwork] [reading-I] [reading-II]
  
• 

  Generative Learning Algorithms: Continued
  tl;dr: Gaussian discriminant analysis, linear decision boundary, quadratic decision boundary, Naive Bayes Classifier, Smoothing
  [boardwork] [reading-I] [reading-II]
  
• 

  Support Vector Machines (SVMs) -- I
  tl;dr: Basics of coordinate geometry, Lagrange Multipliers, KKT Conditions
  [scribe] [boardwork] [KKT - Lecture Note] [reading-I]
  
• 

  Support Vector Machines (SVMs) -- II
  tl;dr: Primal and Dual Problems, Objective function of SVM, Optimization of SVM
  [scribe] [boardwork] [reading-I]
  
• 

  Support Vector Machines (SVMs) -- III
  tl;dr: tl;dr: Kerner tricks -- Linear, Polynomial, RBF
  [boardwork] [reading-I] [reading-II]
  
• 

  Support Vector Machines (SVMs) -- IV
  tl;dr: Kerner tricks -- SVM Soft Margin, bias and variance
  [scribe] [reading-I]
  
• 

  Bias-Variance Tradeoff
  tl;dr: Mathematical derivation of bias, variance and noise and discuss their trade-off
  [scribe] [reading-I]
  

  Supplementary Reading:

  • Understanding the Bias-Variance Tradeoff
• 

  Ensemble learning
  tl;dr: Bagging, Boosting, Random Forest
  [scribe] [reading-I] [reading-II]
  
• 

  Boosting
  tl;dr: Weak learner, Boosting reduces bias, Anyboost and Adaboost algorithms
  [scribe] [reading-I] [reading-II] [reading-III]
  
• 

  Instance-based learning
  tl;dr: tl;dr: kNN, error analysis, bias and variance
  [scribe] [boardwork] [reading-I]
  

  Supplementary Reading:

  • Tom Mitchell
• 

  Dimensionality Reduction
  tl;dr: Principal component analysis, Singular value decomposition, Linear Discriminant analysis
  [scribe] [PCA] [LDA] [SVD]
  
• 

  Dimensionality Reduction -- Part II, Introduction to Neural Networks
  tl;dr: Linear Discriminant Analysis, History of Neural Networks
  [boardwork]
  
• 

  Machine Learning, Everywhere! - Guest Lecture by Dr. Sumit Bhatia
  tl;dr: Overview of ML’s importance in Adobe’s operations, Case studies showcasing how ML drives product innovation and enhancement, impacting everyday life from space missions (Chandrayaan) to online deliveries.
  
  
• 

  Perceptron and Multi-layer Perceptron
  tl;dr: MPNet, Perceptron, Multi-layer Perceptron, Activation Function, Backpropagation
  [reading-I]
  
• 

  CNN, Intro to RNN
  tl;dr: Convolutional Network Networks, Recurrent Neural Networks
  [CNN Slides] [RNN Slides] [CNN Reading]
  

  Supplementary Material:

  • RNN Video
• 

  RNN (conclusion), Transformer
  tl;dr: BPTT, issues with RNNs, LSTM, Attention
  [RNN Slides] [LSTM Slides] [Attention Slides]
  

  Supplementary Material:

  • RNN Video
  • LSTM Video
  • Attention Video
• 

  Transformer (Conclusion)
  tl;dr: Self-attention, masked self-attention, encoder-decoder transformer
  [Slides-I] [Slides-II]
  

  Supplementary Material:

  • Transformer-I
  • Transformer-II