# The Matrix Calculus You Need For Deep Learning

**Authors:** Terence Parr, Jeremy Howard  
**arXiv:** [1802.01528](https://arxiv.org/abs/1802.01528)  
**Date:** 2018-02-05  
**Summary page:** https://www.summarizepaper.com/en/paper/1802.01528v1/

## Key points

- "The Matrix Calculus You Need For Deep Learning" by Terence Parr and Jeremy Howard explains matrix calculus essential for training deep neural networks
- Assumes no prior math knowledge beyond calculus 1, provides links to refresh necessary concepts
- Not a prerequisite for learning and applying deep learning techniques, targets readers familiar with neural network basics seeking deeper understanding of mathematics
- Encourages revisiting previous sections and working through examples if difficulties arise
- Additional support available through the Theory category on forums.fast.ai
- Includes a reference section summarizing key matrix calculus rules and terminology discussed throughout
- PDF version optimized for mobile/web viewing is available at http://parrt.cs.usfca.edu/doc/matrix-calculus/index.html

## Summary

In "The Matrix Calculus You Need For Deep Learning," authors Terence Parr and Jeremy Howard provide a comprehensive explanation of the matrix calculus required for understanding the training of deep neural networks. The paper assumes no prior math knowledge beyond calculus 1 and offers links to refresh necessary concepts as needed. It emphasizes that this material is not a prerequisite for learning and applying deep learning techniques but targets those familiar with neural network basics seeking a deeper understanding of the underlying mathematics. The authors encourage readers to revisit previous sections and work through examples if they encounter difficulties, with additional support available through the Theory category on forums.fast.ai. The paper concludes with a reference section summarizing key matrix calculus rules and terminology discussed throughout. Additionally, a PDF version optimized for mobile/web viewing is available at http://parrt.cs.usfca.edu/doc/matrix-calculus/index.html.