Using deep learning to label feedback messages
In this blog post I write about how Nodes uses deep learning to classify user-generated feedback messages across our digital solutions. We do this to create ...
archive
What is the precision matrix?
Refreshing my understand of the meaning of the precision matrix.
Extracing city blocks from a graph
Finding blocks and defining neighborhoods in city data is surprisingly non-trivial. Fundamentally, it amounts to finding the smallest set of rings (SSSR), wh...
Running Vowpal Wabbit in Docker
Install and run Vowpal Wabbit in a Docker container.
ML on Rails
Learn to build a simple Rails web application that calls a machine learning model and exposes the result with a click of a button.
A tutorial on the free-energy framework for modelling perception and learning
Based on K. Friston and following R. Bogacz I shortly summarise the former, and solve the programming exercises given in the paper.
Non-ergodicity in a simple coin game
Based on a seminal talk by Ole Peters, I illustrate the fundamental difference between an ergodic and non-ergodic process using a simple coin game.