This repository contains the notebooks and materials used in the Pattern analysis and Machine Intellegince course. The lecture and the practical assignments will be uploaded on Mondays. The solutions to the exercises will be uploaded on Thursday. The Friday session will be reserved for questions and discussion about the previous exercises. We will present some talks about current research trends that relate to the practiced material. These talks will be done by various researchers and serve as an opportunity for the students to get familiar with current literature for their projects.
Course Schedule:
17.04 - week 1: General introduction to machine learning and course utilities. + Classification with Naive Bayes + Logistic regression and gradient descent Practical assignments: (optional) python review notebook. Spam message identification + Gradient descent variants and Titanic survivor prediction
24.04 - week 2: Introduction to neural networks: multi-layer perceptron (MLP), convolutional neural network (CNN). + ML Frameworks Practical assignment: (Optional): MLP classification of fashion images with pure Numpy + Fashion image classification and reading traditional Japanese character PyTorch.
01.05: week 3. Holiday Practical assignment: No Practical Assignment
08.05 - week 4: Random Forests. Practical assignment: Revisiting Titanic survivor prediction + an additional challenge (San Francisco Crime).
15.05 - week 5: Neural sequence models, recurrent neural networks. Practical assignment: Shakespeare poetry text generation.
22.05 - week 6: Transformers and Attention Mechanisms ( Practical assignment: Text Generation with Transformers
29.05 - week 7 Lecture: Holiday Practical Assignment: No Practical Assignment
05.06 week 8: Lecture: Time Series Forecasting with NN vs Classical Methods. Practical assignment: TBD
12.06 - week 9: Lecture: Introduction to unsupervised learning. Clustering with k-means. Principal and independent component analysis (PCA and ICA). Practical assignment: Finding clusters in generated distributions and signal source separation.
19.06 - week 10: Unsupervised neural networks: representation learning through variational auto-encoding (VAE). And Generative adversarial networks (GAN). Practical assignment: Revisiting fashion and Kuzushiji for unsupervised pre-training and image generation + Face generation using GANs
26.06 - week 11: Self Supervised Learning: Contrastive Learning and similar methods. Practical assignment: SimCLR and BYOL for image classification.
03.07 - week 12: Lecture :Classic tabular q-learning. + Deep reinforcement learning, QNN Practical assignment: (Optional) Cart pole balancing. + Taxi driver with DeepRL
10.07 - week 13: Meta-learning. Practical assignment: neural architecture search using the reinforce algorithm.