Joseph Van Name

Ph.D. in Mathematics. I have been posting visualizations of machine learning training and mathematics. My focus on this channel has been pseudodeterministic machine learning, Laver tables, and whatever I find interesting at the moment. We say that a machine learning model is pseudodeterministically trained (or p-deterministic) if whenever we train it multiple times, we obtain the same trained model. Many pseudodeterministic machine learning models have been featured on this channel, and as of 2026, I am developing pseudodeterministic deep learning models.

Descriptions of the algorithms involved here can be found at circcashcore.org or at github.com/jvanname

I am available to consult your business on the use of AI (at a rate between 100 USD/hr and 150 USD/hr), so feel free to contact me if you want to use AI more safely and effectively. You can also support my work on AI safety and interpretability research at https://github.com/sponsors/jvanname.

Please repent of your sins.