UNDERSTANDING DEEPMIND'S SORTING ALGORITHM

Summary from the TLDR Web Dev newsletter

A few days ago, DeepMind published a blog post talking about how they discovered tinier kernels for sorting algorithms. They did this by taking their deep learning wisdom, which they gained by building AlphaGo, and applying it to the discipline of superoptimization. This article explains the optimization that was made more clearly compared to the actual blog post that was published.