Mathematics for Machine Learning. Companion website towards the written book”Mathematics for Machine Learning”.

Copyright 2020 by Marc Peter Deisenroth, A. Aldo Faisal, and Cheng quickly Ong. Posted by Cambridge University Press.

We penned a novel on Mathematics for Machine training that motivates individuals to discover mathematical ideas. The guide just isn’t designed to cover advanced level device learning strategies since there are already an abundance of publications achieving this. Rather, we seek to give you the necessary skills that are mathematical read those other publications.

The guide can be acquired at posted by Cambridge University Press (posted April 2020).

We split the guide into two parts:

  • Mathematical fundamentals
  • Example device learning algorithms that make use of the foundations that are mathematical

We aimed to help keep this guide fairly short, therefore we don’t protect every thing.

Table of articles

Component We: Mathematical Foundations

  1. Motivation and introduction
  2. Linear Algebra
  3. Analytic Geometry
  4. Matrix Decompositions
  5. Vector Calculus
  6. Likelihood and Distribution
  7. Continuous Optimization

Component II: Central Machine Training Issues

  1. Whenever Versions Meet Data
  2. Linear Regression
  3. Dimensionality Decrease with Principal Component Analysis
  4. Density Estimation with Gaussian Mixture Versions
  5. Category with Help Vector Machines

Report errata and feedback.

Any dilemmas you raise now might not allow it to be to the printed version, but we will keep an updated PDF around (and also the errata).


PDF associated with the guide

This variation is considered the most up-to-date type of the guide, for example., we carry on fixing typos etc.

Instructor’s manual solutions that are containing the workouts (could be required from Cambridge University Press)

Errata on overleaf

PDF associated with the printed guide

This variation is equivalent (modulo formatting) aided by the printed form of the book. GitHub dilemmas beginning 433 aren’t most notable variation.

Methods to workouts

  • Instructor’s manual solutions that are containing the workouts (could be required from Cambridge University Press)
  • Extra workouts (with solutions)


  • Jupyter notebook tutorials (for learning)
    1. Linear Regression
    2. PCA
    3. Gaussian Mixture Versions
    4. SVM (work with progress)
  • Jupyter notebook tutorials (solutions)
    1. Linear Regression
    2. PCA
    3. Gaussian Mixture Versions
    4. SVM (work with progress)
  • NeurIPS-2020 tutorial on differentiation and integration

Outside resources

Other folks have actually produced resources that support the materials in this guide.


‘This guide provides great coverage of most the essential mathematical ideas for device learning. I’m looking towards sharing it with pupils, peers, and anyone thinking about building a understanding that is solid of fundamentals.’ Joelle Pineau, McGill University and Twitter

‘The industry of device learning is continuing to grow considerably in the last few years, with an spectrum that is increasingly impressive of applications. This text that is comprehensive the main element mathematical ideas that underpin modern machine learning, with a focus on linear algebra, calculus, and likelihood concept. It’s going to show both that is valuable a tutorial for newcomers to your industry, so that as a reference text for machine learning researchers and designers.’ Christopher Bishop, Microsoft Analysis Cambridge

‘This guide provides a lovely exposition associated with math underpinning modern device learning. Strongly suggested for anybody wanting a shop that is one-stop obtain a deep comprehension of device learning foundations.’ Pieter Abbeel, University of California, Berkeley

‘The guide strikes the level that is right of in my situation. Way too many regarding the ML publications have actually a “don’t worry your pretty mind concerning this detail” mentality, or get one other method and overwhelm me personally with detail. Your guide is comprehensive and it has a feeling of simplicity and expanse, however it feels as though I’m able to arrive at the application form component quickly sufficient.’ Sriram Srinivasan

Hosted on GitHub Pages — Theme by orderedlist


Leave a Comment

© // Ristorante & Bar