Start from scratch: machine learning 20161101

Starting today to learn machine learning, mainly in several aspects, is machine learning for my personal several aspects of the promotion is particularly large. Whether it's a financial or an image.

In Finance I need machine learning to reconstruct and evaluate my strategy.

In the image, I am good at pattern recognition, need a lot of data processing, machine learning is the fundamental of big data.

Start today and learn a little every day. Keep going.

The key to machine learning is how to get started and how to reduce dimensions. I have some of the most optimized ideas. The least squares were used before. is also part of the idea of optimization.

The learning materials currently found are videos and handouts from Stanford University's machine learning public course. On the whole, the handout works better.

The key to the present is the habit of planning and writing code. It is still better to use MATLAB and octave to deal with the verification algorithm. It is also possible to use Python for further understanding.

Tell me about the current known situation:

Supervised learning and unsupervised learning

Clustering and classification

Ideas and prototypes for various algorithms

Start from scratch: machine learning 20161101