Goal

This post will provide you with a detailed Machine Learning Roadmap to help you begin your ML adventure. Whether you are a novice or have just begun, you will find basic to advanced level machine learning resources. I, too, used these sites to learn a lot about machine learning.

Problem

If you want to establish a career in AI and ML, you may face many problems and confusions in the beginning because many of us do not know the correct resources and path to take for machine learning. Some of us may end up taking advanced courses too early in the adventure get overwhelmed and give up.

Definition

Machine Learning is an interdisciplinary discipline that employs statistics, probability, and algorithms to learn from data and provide insights that may be utilized to construct intelligent systems. Machine Learning is a mix of Mathematics, Statistics and Computer Science

Foreword

I’m a firm believer that it’s crucial for an ml practitioner to understand both the theoretical math rigorous side and practical sides. The first step should be gaining mastery in python and its data related libraries while also diving in the math prerequisites to build a good foundation to understand ml concepts

Roadmap

Maths

Mathematics is the heart of Machine Learning In my opinion, having a basic understanding of math is required when beginning Machine Learning because every Machine Learning algorithm is based on mathematical notions. If you are a university student, you may already have a basic understanding of mathematics, which is necessary for Machine Learning. Don’t be sad if you aren’t. I’ll guide you through every mathematical resource you’ll need before diving into ML.

Linear Algebra

Linear Algebra is a subfield of mathematics. In Linear Algebra, we investigate vectors, matrices, Symmetric Matrices, Orthogonalization & Orthonormalization, Matrix Operations, Projections, Eigenvalues & Eigenvectors, Vector Spaces, and Norms.

Resources:

• 3Blue1Brown’s playlist
• Khan Academy’s Linear Algebra course
• Gilbert Strang’s lectures on MIT OpenCourseWare
• Gilbert Strang’s book: Introduction to Linear Algebra 5th Edition

Probability & Statistics

ML theory employs probability and statistics: Combinatorics, Probability Rules & Axioms, Bayes’ Theorem, Random Variables, Variance and Expectation, Standard Distributions (Bernoulli, Binomial, Multinomial, Uniform, and Gaussian), Conditional and Joint Distributions, Moment Generating Functions, Prior and Posterior, Maximum Likelihood Estimation (MLE), Maximum a Posteriori Estimation (MAP) and Sampling Methods.

Resources:

• The Seeing Theory project platform
• Probability: For the Enthusiastic Beginner by David Morin
• Introduction to Statistics on Udacity
• Practical Statistics for Data Scientists: 50 Essential Concepts 1st Edition
• An Introduction to Statistical Learning

Calculus

Calculus assists in understanding the behavior and internal workings of various algorithms and functions. It is best to learn calculus to understand how the function behaves and changes over time. Differential and integral calculus, partial derivatives, vector-valued functions, directional gradients, Hessian, Jacobian, Laplacian, and Lagrangian distributions are among the topics covered in calculus.

Resources:

• 3Blue1Brown’s Calculus playlist
• Khan Academy’s Course
• Calculus: Early Transcendentals 8th Edition by James Stewart

Discrete Mathematics

Discrete Mathematics is not usually mentioned in other roadmaps but it’s important to grasping most of computer science related fields It is concerned with the study of countable, discrete components, and its ideas are frequently employed in the development of algorithms for computer science and data science. Knowledge of discrete math ideas may help you grasp algorithms, binary, and general mathematics, which are at the heart of data-driven jobs.

Resources:

• Discrete Mathematics with Applications by Susanna Epp

Python

Python is the industry standard when it comes to data science and machine learning

Resources:

• FreeCodeCamp’s video
• Intermediate Python Programming video by Patrick
• Automate the Boring Stuff with Python

Then you need to learn the necessary libraries for handling data:

• Youtube - Data Analysis with Python - Full Course for Beginners
• Python for Data Analysis You also have to learn Algorithms and Data Structures
• Grokking Algorithms
• Cracking the Coding Interview

Machine Learning

You have now viewed all of the prerequisites for machine learning requirements. It is now time to dive into the fundamentals of machine learning algorithms. As you may know, machine learning is the science of making machines act like humans and make judgments and predictions for various purposes. It is essential to have a basic understanding of ml algorithms.

ML Pipeline Gathering Data Preparing that data Choosing a model Training Evaluation Hyperparameter Tuning Prediction

Resources:

• Andrew Ng’s course on Coursera
• DataCamp’s courses
• The Hundred-Page Machine Learning Book
• Pattern Recognition and Machine Learning
• Hands-On Machine Learning 2nd Edition

This is merely the resources I’m currently using to learn about this field, I am no expert in this. I recommend taking notes while reading and learning using a platform like notion, your future self will thank you a lot for that,