Articles
Mastering Subplot Legends in Matplotlib
A tutorial on how to create shared subplot legends in Matplotlib. Python code provided in a Jupyter notebook.
Scientists Are Now Teaching AI How To Read Human Minds
Artificial Intelligence (AI) can read your mind... if you are hooked up to an fMRI machine and it is trained on how you process visual information.
Robots that can assemble almost anything (featuring my interview on AI for predicting suicide risk)
Robots that can assemble almost anything a step closer. Also the war of words on Wikipedia; attempts at changing the narrative. Finally can AI predict suicide risk?
AI could help predict suicides – but rushing the technology could lead to big mistakes
AI has great potential to save the lives of mental health patients, but care must be taken to ensure it's used safely and correctly.
AI can produce prize-winning art, but it still can't compete with human creativity
People consider creativity to be inherently human. However, artificial intelligence (AI) has reached the stage where it can be creative as well. But could AI actually replace human artists?
Explainable AI for High-Resolution Images
AI can be used to make automated decisions based on high-resolution images, but can we understand those decisions?
Mastering Inset Axes in Matplotlib
Inset axes are a fantastic (and often underutilised) tool in Matplotlib. In this article, I go through the basics of using inset axes, and then give extra details on how to customize and improve your plots.
Three Minute Thesis - 2022 Southampton Engineering Heat Entry
The Three Minute Thesis (3MT) competition celebrates the exciting research conducted by PhD students. This my winning entry to the 2022 Southampton Engineering Heat.
Understanding the Outputs of Multi-Layer Bi-Directional LSTMs
In this short tutorial, I break down the outputs of Multi-Layer Bi-Directional LSTMs, with an example of how to do so in PyTorch.
Can AI solve the world's hardest Go problem?
Igo Hatsuyoron 120 is the hardest Go problem in the world. Let's explore the most recent developments in applying AI to this problem.
The Potential of Gated Linear Networks for Online Learning
A recent publication from DeepMind brings an interesting new take on sample-efficient online learning.
How to create Latex tables directly from Python code
Copying tables of results from the console into a Latex report can be tedious and error fraught — so why not automate it?
What's hot in Multi-Agent Systems?
Multi-Agent Systems research has been a long standing topic in AI, but what are AI researchers in the UK currently working on?
The Biggest AI Risk of the Next Decade is not a Robot Uprising
Grand generalisations about the future impacts of artificial general intelligence overshadow the more pressing issues we face today.
Reproducible Data Science using Kaggle and GitHub Actions
This tutorial demonstrates how to integrate Kaggle with GitHub actions to enable better reproducibility of data science projects.
Your Car May Not Know When to Stop - Adversarial Attacks Against Autonomous Vehicles
Autonomous vehicles present a utopian dream of high-speed, clean and efficient transportation, but the computer vision systems at their core are susceptible to adversarial attacks.
Real Time Image Style Interpolation
My submission to the ICLR Reproducibility Challenge: a re-implementation of "A Learned Representation For Artistic Style" (Dumoulin et al., 2016).