This Person (Probably) Exists
An interesting paper I came across recently that discusses privacy concerns surrounding generative machine learning models.
An interesting paper I came across recently that discusses privacy concerns surrounding generative machine learning models.
I recently attended the 6th Annual Visualization for AI Explainability Workshop, which explored some innovative ways to use visualizations to explain machine learning concepts as well as build trust in ML models.
'Risk' is an appropriately named board game which revolves around battling opposing armies with the aim of invading their territories and eventually conquering the world. The game involves a large amount of strategy and planning but also some luck; battles are ultimately won and lost by the rolls of dice. This post is going to explore the battle mechanic in more detail, starting by calculating the probability of winning in different scenarios and then showing how we can get a deeper insight into these numbers with some data visualisation.
A review of my winning entry from the source code poetry competition 2020
An interactive game to demonstrate how decision trees work. Built using D3.js
The decision tree is one of the first classification algorithms taught in any machine learning course. It is easy to understand and gives a very interpretable output. Despite this, most people don't realise that all tree building algorithms (e.g. CART, ID3, C.45, MARS) are suboptimal.