Intelligent Systems for Pattern Recognition 2023/2024 course repository.
This repository contains all solutions of the assignments that I chose to solve throughout the course:
- Assignment 1: Time series analysis on the Air Pollution Dataset
- Assignment 2: Restricted Boltzmann Machines implementation from scratch, tested on MNIST
- Assignment 3: Denoising and Contractive Autoencoders implementation from scratch, digit generation test on MNIST
- Assignment 4: 8 slides presentation of Masked Autoregressive Flow for Density Estimation - George Papamakarios, Theo Pavlakou, Iain Murray
The complete and detailed notes of the course can be downloaded in the Notes23-24.pdf file, written in collaboration with Francesco Aliprand, list of topics:
- Signal processing (time and frequency analysys)
- Image processing (descriptor and detectors)
- Wavelets
- Probabilistic Machine Learning
- Bayesian Networks
- Hidden Markov Models
- Markov Random Fields
- Bayesian Learning and Variational Inference (with Latent Dirichlet Allocation)
- Sampling
- Boltzmann Machines
- Deep Learning
- Convolutional Neural Networks
- Autoencoder Models
- Gated Recurrent Networks
- Attention - Based architectures
- Neural Memories
- Generative Models
- Explicit Density Learning (Variational AE)
- Adversarial Learning (Generative Adversarial Networks)
- Diffusion Models
- Normalizing Flows
- Deep Graph Networks
- Deep Reinforcement Learning