Random Projections

LLMs, CoT and Implicit Reasoning

Machine Learning

The last few months have been tumultuous for AI, with Deepseek’s R1 demonstrating how Reinforcement Learning can bring out enough reasoning power from smaller models to…

Dimensionless Features

Machine Learning

When dealing with machine learning problems for physical systems the quantities we use as features and targets have dimensionality.

Bayesian Persuasion

Machine Learning

Bayesian Analysis

In this post we explore the connections between standard Machine Learning methods and Bayesian inference. We also show how to build a Bayesian neural network without relying…