Week 3:

Lectures:

Computational Genomics
Uncertainty Modelling
Learning in Autonomous Systems

Not the best week - Operation Recuperation was not what you'd call a roaring success - however it cannot be said that I was totally unproductive: I made a start on my coursework for Autonomous Systems and have made progress in deciphering the background to my Thesis. During this week I will endeavour to post some of this insight and to explain a little about what I've done for the aforementioned coursework. As a taster, here is a basic outline of what the work is about:

Learning in Autonomous Systems - Assignment 1: CVRP

This project is based on finding solutions to an optimisation problem called the Capacitated Vehicle Routing Problem (CVRP). The basic concept is that of a supply depot needing to find routes for its delivery vehicles which will make sure all customers receive the correct (whole number) amount of ordered goods. The 'capacitated' bit means that the vehicles can only carry a fixed amount. The optimisation bit comes from the fact that the depot wants to minimise cost and therefore requires the 'optimal' (ie shortest total route) solution. I will be working on the project from a machine learning point of view, where the aim is for the program I write to perform multiple iterations of a given algorithm to hone in on a solution which is approaching optimal. It's a while since I've done programming of this type, and I may be a little rusty on implementing algorithms like this in code form, but I will report progress here, as and when it occurs.