ES 335 IIT Gandhinagar

Home Lectures Assessment FAQ Slack

Lectures

Each lecture lists its slides and reading first, then any recording, then its notebooks. Slides and notebooks come from ml-teaching, which also lists all slides and all notebooks.

SlidesFull recordingNotebookReading

Foundations1–2Trees, model fit and ensembles3–7Linear models and optimisation8–14Classification15–18Neural networks19–26Unsupervised learning, neighbours and reinforcement learning27–29Support vector machines30–33

Foundations

Lectures 1–2

·	Prerequisite test An open-book, open-internet test, required to take the course.	Prerequisite test
Lecture 01	Introduction and logistics	Slides Notebook: transcription and local LLMsNotebook: segmentation and localisationNotebook: prompt to image
Lecture 02	Conventions, metrics, classification and regression	Slides Notebook: rule-based vs MLNotebook: confusion matrixNotebook: dummy baselines

Trees, model fit and ensembles

Lectures 3–7

Lecture 03	Decision trees I	Slides Notebook: entropyNotebook: decision trees, real outputNotebook: decision trees, real input and discrete output
Lecture 04	Decision trees II; bias and variance	Slides Notebook: bias and varianceNotebook: decision trees, overfitting and underfitting
Lecture 05	Tutorial: NumPy, Pandas and classes; cross-validation	Slides: cross-validation Notebook: NumPy, Pandas and entropyNotebook: classes and treesNotebook: cross-validation and hyperparameter tuning
Lecture 06	Random forests	Slides Notebook: random forest feature importance from scratch
Lecture 07	Tutorial: random forest feature importance, meshgrid and contour plots	Notebook: random forest feature importance from scratchNotebook: meshgrid

Linear models and optimisation

Lectures 8–14

Lecture 08	Linear regression	Slides
Lecture 09	Linear regression: basis functions	Slides Notebook: basis functions
Lecture 10	Linear regression: geometric interpretation, multicollinearity and dummy variables	Slides Notebook: geometric interpretationNotebook: multicollinearity and dummy variables
Lecture 11	Gradient descent	Slides Notebook: Taylor seriesNotebook: contour plots and gradient descent
Lecture 12	Gradient descent (continued)	Slides Notebook: Taylor seriesNotebook: contour plots and gradient descent
Lecture 13	Movie recommendation and matrix factorisation	Slides Notebook: movie recommendation
Lecture 14	Regularised linear regression: ridge and lasso	Slides: ridge regressionSlides: lasso regression

Classification

Lectures 15–18

Lecture 15	Logistic regression	Slides Notebook: why use logitsNotebook: logistic regression in PyTorch
Lecture 16	Logistic regression (continued)	Slides Notebook: why use logitsNotebook: logistic regression in PyTorch
Lecture 17	Tutorial: convexity	Slides
Lecture 18	Weighted linear regression; convexity of the cross-entropy loss and IRLS	Slides: weighted linear regressionSlides: cross-entropy convexity and IRLS

Neural networks

Lectures 19–26

Lecture 19	Precision–recall curves; multilayer perceptrons I	Slides Notebook: precision–recall curves
Lecture 20	Multilayer perceptrons I (continued)	Slides
Lecture 21	Multilayer perceptrons II	Slides
Lecture 22	Autograd	Slides Notebook: automatic differentiation
Lecture 23	Next-token prediction	Slides Notebook: generating names
Lecture 24	Convolutional neural networks I; 1D CNNs	Slides: CNNSlides: 1D CNN Notebook: 1D CNNNotebook: edge detectionNotebook: LeNet
Lecture 25	Convolutional neural networks II	Slides: CNN
Lecture 26	Tutorial: object detection and setting up Kaggle

Unsupervised learning, neighbours and reinforcement learning

Lectures 27–29

Lecture 27	Unsupervised learning	Slides
Lecture 28	K-nearest neighbours and approximate KNN	Slides: KNNSlides: approximate KNNReading: parametric vs non-parametric models (MLSS, Ghahramani)Reading: Machine Learning MasteryReading: Cross Validated answerReading: Sebastian Raschka's FAQ Notebook: parametric vs non-parametricNotebook: curse of dimensionality
Lecture 29	Reinforcement learning	Slides Notebook: reinforcement learning, part 1Notebook: reinforcement learning, part 2

Support vector machines

Lectures 30–33

Lecture 30	Support vector machines I	Slides Notebook: SVM
Lecture 31	Support vector machines II: kernels	Slides (from SVM I) Notebook: primal and dualNotebook: kernels
Lecture 32	Constrained optimisation	Slides: constrained optimisation 1Slides: constrained optimisation 2 (self-study) Part 2 recording from an earlier offering
Lecture 33	Support vector machines: soft margin	Slides

Slides are the PDF handouts used in class. Quiz days are listed under Assessment.