Theory of Intelligence
MACHINE LEARNING•Unsupervised Learning•Dimensionality Reduction•1.5 Random Projection, LLE & t-SNE

Introduction

Three more dimensionality-reduction tools: random projection for very high dimensions, LLE for unrolling manifolds, and t-SNE for beautiful 2-D visualisations.

Introduction

PCA is the workhorse, but other tools are better for particular jobs:

ToolBest for
Random projectionextremely high-dimensional data, when speed matters most
LLEunrolling curved manifolds (like the Swiss roll)
t-SNE (and UMAP)2-D/3-D visualisation of clusters

Discuss theoretical concepts, mathematical formulas, and architectural foundations.

Supports inline code with `backticks`
No comments yet. Be the first to start the discussion!