How to get started with Machine Learning?

This is most basic question people ask is how to get started with machine learning? Here are some basic guidelines, no specific course but generalised list of topics which one should cover if he/she wants to be in machine learning.

The road-map is divided into 6 sections. They are:

1. Mathematics
2. Python/R
3. Machine Learning Algorithms
4. Deep Learning Algorithms
5. Computer Vision
6. Natural Language Processing

Mathematics

• Linear Algebra
• Statistics
• Calculus
• Probability
• Numerical Optimisation

Python/R

• Basic Python/R
• Numpy, Pandas, sci-kit learn, Matplotlib, nltk

Machine Learning Algorithms

• Regression, Classification, Clustering, Gradient Descent, Optimisations
• Bayesian Methods, Time Series, Ensemble Algorithms
• Libraries: sci-kit learn, graphlab.

Deep Learning Algorithms

• Artificial Neural Networks (ANN), Activation Functions, Dropout
• Convolution Neural Networks (CNN)
• Recurrent Neural Networks (RNN)
• Gated Recurrent Units (GRUs) Long-short term memory Networks (LSTMN)
• Deep Belief Networks
• Transfer Learning and other advanced concepts such as R - CNN, F - RCNN
• Frameworks: Tensorflow, Keras, Pytorch, Caffe, gym by openai.

Computer Vision

• Open-CV
• Pattern Recognition
• Segmentation
• Object detection

Natural Language Processing

• Tokenization, Stemming and Lemmatizing
• POS (Part of Speech)
• NER (Named Entity Recognition)
• TF-IDF
• word2vec

These all are pure mathematical topics, implement all of them at least once.