Blog

This post explores how making AI models more transparent can build trust and improve decision-making. It introduces Linearised Additive Models (LAMs), which enhance the clarity of logistic models without sacrificing performance.

In this post we analyse the new betting game, last man standing. We'll look at strategies for winning the game under certain assumptions, using techniques such as simulated annealing and convex programming.

This post is a tongue-in-cheek look at using complexity theory to justify poor decision-making in our day to day lives.

We investigate using a deep neural network to classify audio data as a 'banger' or 'not_a_banger', inspired by *Parks and Recreation*'s Tom Haverford.

The companion notebook to is-it-a-banger, this shows how to create the dataset used in the article.