I used to teach AI & ML for media, art & design: PerttuHamalainen/MediaAI: Aalto University's Intelligent Computational Media course (AI & ML for media, art & design)

However, I killed the course as one can increasingly get better results with commercial tools than tweaking and finetuning models oneself. Instead, I’m now focusing more on teaching advanced game programming and technical art, with some AI & ML material sprinkled in.

Meanwhile, I’ll keep posting significant new papers etc. here.

Note that this section is very much under construction, waiting for me to clean up and migrate stuff from my private iCloud notes.

Table of contents

LLMs

Kolibri, a sovereign European LLM, with a comprehensive tech report of the training recipe etc.

October 4, 2026

Benchmarks are not SotA, but still ok, and the 184 page tech report is more comprehensive than those shared by closed labs or the papers from Deepseek etc. A very nice ”all-in-one” document for leaning the details of modern LLM architecture and training.

https://aleph-alpha.com/downloads/tech-report.pdf

The only major problem is that Kolibri is not vision-enabled.

Symmetry and geometric structure in how neural networks represent data

September 6, 2026

A brilliant talk by Max Tegmark that packs in a ton of examples of how LLMs and other machine learning models discover efficient latent representations that help them generalize and reason:

https://youtu.be/nRrt7AczYV4?is=Km059J98lc-31yq_