Deep Learning AI #1

deep-learning-ai.md

Course info

• Motivation: Deep learning in vision, speech, text, robotics.

• Deep learning courses:

  • Andrew Ng: Deep Learning
  • fast.ai: Practical Deep Learning For Coders
  • Stanford: CS231n Convolutional Neural Networks for Visual Recognition
  • Oxford: Machine Learning

• Flipped class-room format: watch videos and attempt exercises at home, meet to discuss.

• Start early to go on through the course, but meet officially in mid-January.

• Slack channel, deep-learning-course, for internal discussion.

• Coursera overview: how to audit the course and introduction to materials (videos, assignments, discussion forum).

Pre-requisites

From the FAQ

Pre-requisites:

• Very basic programming skills (i.e. ability to work with dictionaries and for loops).
• Familiarity with basic machine learning (how do we represent a dataset as a matrix, etc.).
• Familiarity with the basic linear algebra (matrix multiplications, vector operations etc.).

• Advice: Skim the following resources; don't go too much into depth; revisit as needed.
• The recommended resources are marked with R.

Programming

Python3

Tools and libraries:

• jupyter -- interactive computing.
• numpy -- package for scientific computing.
• matplotlib -- plotting library.
• sklearn -- machine learning toolbox.
• keras -- deep learning library.

General Python:

• R Python3 documentation
• Learn x in y minutes

Scientific computing:

• R A Crash Course in Python for Scientists
• R NumPy for MATLAB users

Exercises or projects:

• R Check the introductory assignment in the course
• R ETTI programming exam
• Other: Katas; Project Euler; Python challenge; Project ideas.

Machine learning

• R For a refresher check the first couple lectures from Andrew Ng's Machine Learning course.

• Other introductory resources:

  • Hal Daume III -- A Course in Machine Learning book
  • Iain Murray -- Introduction to Machine Learning video
  • Pedro Domingos -- A few useful things to know about Machine Learning article
  • A Visual Introduction to Machine Learning visualization

Linear algebra

• R Stanford CS229 Machine Learning: Linear Algebra Review and Reference
• Iain Murray's cribsheet
• Matrix cookbook

---

Feel free to add other recommendations.