2 min to read

Resources to learn ML

Prerequisites for learning ML

Featured image

Introduction

Machine Learning is a broad area. In this post, I will be sharing some resources that proved quite helpful to me (Although I am no one in this area but if you consider me as atleast an okay student, believe me these are great resources) I have tried to keep the list concise so that you can get started quickly and easily

Building the foundation

First you need to learn probability, algebra and statistics to dive deeper into this area

  1. Probability lectures MIT by Prof Tsitsiklis

    These lectures are available on youtube as well

  2. Famous 18.06 course by Prof. Gilbert Strang
  3. Apart from these Linear Algebra by gilbert strang is also very good.

  4. Schaum’s Introduction of Probability and Statistics by Prof Lipschutz and Prof. Schiller

    For statistics and basic probability this is quite good as a stand alone book. Also it contains many solved problems which is good for a newbie as well as advanced folks

  5. Atleast basic Python from first 5-6 chapters “fluent python” book. However this book is great and you should read it completely. Atleast preactice some basic DSA skills at interviewbit.com

  6. Learn Pandas from datawars (As of Nov 2023 most of it is free )

Diving into ML

For ML, there are many lectures, but they are good only for applied works. If you want to delve into applied as well want to know inside out of algorithms, I will strongly recommend reading following books once you have gone through resources in above section.

  1. Pattern Recognition and Machine Learning by Bishop
  2. Probabilitic Machine learning by Murphy

  3. Machine Learning: An algorithmic Perspective by stephen marsland

    This book is quite good for doing a quick revision before interviews, exams etc. once you have read either of above two books

Diving into Deep Learning

  1. Prof. Mitesh Khapra’s Deep Learning Lectures

    This is once stop solution to learning deep leaening with all of the matematical as well as implementaion and insights needed for this area. It covers almost all of the topics. Also this is once of the most comprehensice lectures with visualisations that I have found.

  2. You also need to learn one of the Deep Learning libraries for coding needed. You can learn Tensorflow or Pytorch. Learning curve for pytorch is slightly easier than tensorflow but you can learn any one or even both of them. Goto Pytorch and Tensorflow website they already have lots of tutorials on how to begin with it.

Diving into Reinforcement Learning

To be updated soon

Master a particular area

This section contains resources to master a specific area like NLP, computer vision, PGM, etc

To be updated Soon