Use this course to learn at degree level how convolutional neural networks are used for visual recognition
Step-by-step guide
prereq CS229 Machine learning ML, formulating cost fn, taking derivatives,
performing optimisation w/ gradient descent- prereq CS109 basic probability and stats
- get PDF slides and video mpeg4 URLS from http://cs231n.stanford.edu/slides/2017/
- http://cs231n.github.io/ includes assignments
- demos http://vision.stanford.edu/teaching/cs231n-demos/linear-classify/
you should know basic math : linear algebra, partial derivatives, integration, matrix multiplication etc
It is helpful to learn "convolution and correlation for digital signal processing" to understand what the convolution math matrix realy do,
Related articles
silvestro courses
------------------
http://web.stanford.edu/class/cs231m/spring-2014/papers.html
http://web.stanford.edu/class/cs231a/course_notes.html
https://github.com/kenjihata/cs231a-notes