Latest Posts

If you work in data/statistics/DS/ML/AI and are active on LinkedIn you’ve probably seen this image. It typically recieves a large number of positive comments. This makes sense, data by itself isn’t informative without a layer of interpretation. However, without a valid model of the world, ‘telling a story from data’ typically results in an erroneous post-hoc rationalisation of events obtained from noisy, incomplete data that is not representative of the wider population

The current hype around Crypto and web3 is absolutely insane. There’s very little probability we do not look back at this craze as being even sillier than what occured during dot-com bubble.

If you’re somewhat technologically aware and haven’t been living under a rock for the last year, you’ve probably heard of NFT’s, perhaps specifically that one NFT sold for $69 million USD.

Typical bandits are only capable of adapting to new data that are being observed, they are not ‘really’ adapting to changes in the environment. Bandit methods typically assume stationarity and that the environment or underlying phenomenon we are trying to understand doesn’t change in time!

Some of the most popular colour maps are difficult to interpret correctly and are terribly misleading. We demonstrate some of these issues with some simple examples and suggest superior alternatives.

In this post we will explore how playing positive expected value games is not always a good idea, even if you can play them repeatedly. This runs counter to common beliefs regarding expected values.

A passage from Neal Stephenson’s The Diamond Age made me think that perhaps in the future we might be more honest what we now label and market as artificial intelligence.

I’ve written an R package implementing DeepMind’s multidimensional Elo rating approach for evaluating agents. The mELO rating system has the desirable property of being able to handle cyclic, non-transitive interactions (meaning it can handle rock-paper-scissors style dynamics). It is also better behaved in the presence of redundant copies of agents or tasks.

I’ve produced some R code to rapidly generate beautiful high resolution Clifford attractors.