liveBook · Manning

Random rotation

A random rotation is an orthogonal transformation selected randomly to change the orientation of vectors while preserving their lengths and angles.

How 2 Manning books explain it

Quantization and Fast Inference

A random rotation is an orthogonal transformation selected randomly to change the coordinate orientation of numerical vectors.

How it is used
Redistributes normalized KV-cache vector coordinates before scalar quantization in TurboQuant.
How deep it goes
Explains orthogonality, the resulting coordinate distribution, and the quantization role, then shows matrix construction in PyTorch.
Examples
Uses a 128-dimensional attention-head vector and a card-shuffling analogy to illustrate coordinate redistribution.
What you can do after
Apply it by generating an orthogonal matrix with seeded PyTorch QR decomposition and rotating normalized vectors.

Read how Quantization and Fast Inference explains random rotation →

Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI

A random rotation is a reversible transformation that applies a randomly selected orthogonal rotation to input feature vectors.

How it is used
Uses reversible feature transformations to compare model behavior and expose how orientation affects tree-based learning.
How deep it goes
Defines the matrix properties and explains model trade-offs, but does not provide implementation steps or derivations.
Examples
Uses rotated tabular data to contrast gradient-boosted trees with deep neural networks.
What you can do after
Apply a reversible rotation before modeling to test orientation sensitivity and compare tree-based with deep-learning approaches.

Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains random rotation →

Related concepts

Questions about random rotation

What is random rotation?

A random rotation is an orthogonal transformation selected randomly to change the orientation of vectors while preserving their lengths and angles.

Which Manning books explain random rotation?

Quantization and Fast Inference and Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI.

What is the difference between random rotation and gradient boosting?

A random rotation is an orthogonal transformation selected randomly to change the orientation of vectors while preserving their lengths and angles. Gradient boosting is an ensemble-learning technique that sequentially adds predictive models fitted to reduce a loss by following its gradient.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.