The aim of this project is to explore different methods of probabilistic graph generation and walk generation using the igraph library in R. Different properties of graphs are examined and methods of graph generation are compared. By developing practical experience using R and the igraph library, we lay the foundations for better understanding future topics in other projects of the course and review fundamental graph theory.
The aim of this project is to study various properties of social networks which include network’s structural properties, connectivity, degree distribution, personalized networks, community structure, and neighbour based measures. In this project, we first explored the undirected social networks of Facebook and then explored the directed social networks of Google+.
Reinforcement learning is one of the main paradigms of machine learning, alongside supervised and unsupervised learning. Reinforcement learning involves are well-defined and complex interplay of an agent, an environment and actions. The goal of this project is to understand the fundamentals of reinforcement learning by working with its key concepts using two toy examples, which are explored and discussed in detail in this report. In the first part of the project, we will learn the optimal policy of an agent navigating in a 2D environment. We will implement the Value iteration algorithm to learn the optimal policy. In the second part of the project, we will explore the application of IRL in the context of apprenticeship learning.
Graphs are a ubiquitous representation of data and relationships that have applications in all sorts of industries. A reason that graphs are such useful tools is that there are many graph algorithms that allow the extraction of all kinds of information and insight from graph structures. The focus of this project is to work with graph algorithms on real data made available by Uber to develop useful applications. We tackle a variety of problems by implementing algorithms and approximations of algorithms to cement our understanding of some of the most popular graph algorithms. In the first part of the project, we consider a particular graph which models correlations between stock price time series. In the second part, we analyse traffic data on a dataset provided by Uber.