Aidan Cooper
The senior-most engineer I worked with at Past Company had exacting standards when it came to the low-level craft of writing code. Let's call him Tight Typing Tim. Tim kept a close eye on pull requests across the entire codebase, and would consistently call out any
Introducing PGN2FEN — a benchmark for evaluating language models' ability to understand and transcribe chess game move sequences.
Open and private models are becoming more similar than they are different
Building your AI applications around open source models can make them better, cheaper, and faster
The how, why, power, and pitfalls of constraining generative language model outputs
Necessity was the mother of invention. Now, an abundance of cheap storage and compute makes for data anarchy.
In a time of AI prosperity, the life sciences are at risk of being left behind
The how and why of Shapley value approximation, explained in code
What happens to segregated communities as people increasingly seek diversity?
In this article, we will explore how Shapley values work - not using cryptic formulae, but by way of code and simplified explanations
Tree-based models aren't just highly performant - they offer a host of other advantages
This post highlights four common pandas anti-patterns and outlines a complementary set of techniques that you should use instead.
Supervised clustering is a powerful technique that uses SHAP values to identify better-separated clusters than conventional clustering approaches
The empirical utility of some fields of machine learning has rapidly outpaced our understanding of the underlying theory - but we're not sure why.
With interpretability becoming an increasingly important requirement for machine learning projects, there's a growing need for the complex outputs of techniques such as SHAP to be communicated to non-technical stakeholders.