Blog

In the middle of a flurry of work on relaxed semantics, Henzinger et al. published the paper Quantitative Relaxation of Concurrent Data Structures in 2013. Its primary contribution is ironing out a framework for the theory of relaxed semantics. The idea is to allow incorrect transitions between states in the data structure specification while being able to associate different paths or transitions with different costs. Furthermore, they create a relaxed out-of-order stack. Here we will concisely but informally present the theory, and in the end, quickly outline the stack.

I don't think I will ever completely understand Rust lifetimes. Every time I think I get them, they come right back and hit me in the face. The other day I wanted to implement scoped variable bindings for my new programming language Zote, and it took me several hours to figure out why it did not work as I expected. So let's talk a bit about lifetimes, for both of our sakes.

A couple days ago when I checked Samuel's blog, as I do every now and then, I noticed that he had published a new post a whole week ago. Imagine my horror when I realized I was a whole week late to this life changing information! I knew then that it was time to look into RSS and web feeds to find out how I can keep track of updates for things such as this blog. So today I will share my findings with you, so you also can follow Samuel!

Decision trees have been around for several decades as a form of machine learning to mainly classify data. From my own experiences at school and Axis, I saw decision trees as simpler and more explainable when compared to neural networks, but also with less potential. However, it seems that they are used a lot in data mining, where the learning has to be done online. So here we will give a short introduction to decision trees and their online versions.

Welcome to this post and my website in general! Here I aim to ramble a bit about the site and my ideas for what I want to do with it. Let's get rambling :D