Concept map

How the posts relate. Arrows point from a post to the ones that build on it; dashed lines join parts of a series, faint lines join posts that share a tag. Hover to trace neighbours, click to open a post, drag to rearrange, scroll or pinch to zoom.

Reading paths

Every post with a suggested prerequisite, oldest first. Shared tags are browsable from the tag index.

BERT and fine tunning read first: Implementing the Transformer in Python
Understanding Cover's Universal Portfolio read first: Probability Paradigms
Implementing from scratch a Hopfield Network read first: Hopfield Networks for dummies
Implementing the Kalman Filter in Python read first: Probability Paradigms
Having fun with decoding and optimization read first: Implementing the Transformer in Python
Encoder vs Decoder vs EncoderDecoder Architectures read first: Implementing the Transformer in Python
Understanding and Implementing RAG (Retrieval-Augmented Generation) read first: Encoder vs Decoder vs EncoderDecoder Architectures
Diffusion models or Autoencoders? read first: A comprehensive list of different types of VAEs
DDIM vs DDPM read first: Diffusion models or Autoencoders?
Some common divergences read first: Probability Paradigms
Quantization of LLMs read first: Encoder vs Decoder vs EncoderDecoder Architectures
A comprehensive list of different types of VAEs read first: Some common divergences
Temporal Graph Networks: A Deep Dive into Dynamic Graph Learning read first: Graph Convolutional Networks
The Platonic Representation Hypothesis: Are AI Models Converging on Universal Representations? read first: Self Supervised Learning
Scarf: Self Supervised Learning for Tabular Data read first: Self Supervised Learning
Flow Matching, a short overview read first: Diffusion models or Autoencoders?
Diffusion models for time series read first: Diffusion models or Autoencoders?
Classifier free diffusion guidance read first: DDIM vs DDPM
Deepseek, an overview and quick notes read first: Mixture of Experts
Group Relative Policy Optimization (GRPO) read first: Deepseek, an overview and quick notes, Simple Reinforcement Learning Example
Mechanistic Interpretability - Some concepts read first: Implementing the Transformer in Python, Sparse Autoencoders
Mixture of Experts read first: Encoder vs Decoder vs EncoderDecoder Architectures
From Entropy to Epiplexity read first: Kolmogorov Complexity, Some common divergences
H-Neurons: Hallucination at the Neuron Level read first: Mechanistic Interpretability - Some concepts