- #USED AN INTRODUCTION TO STATISTICAL LEARNING HOW TO#
- #USED AN INTRODUCTION TO STATISTICAL LEARNING OFFLINE#
But from my one-on-ones with mentors and an awareness of the fact that there are books and classes on all of these topics individually it’s worth recognizing that this is very much a surface level exploration of the topics. They went a long way towards motivating the concepts, giving you the intuition, and allowing you to solve real problems. I don’t know enough to know exactly how shallow these chapters are. The book does go on to more advanced topics like tree based modeling, support vector machines, and unsupervised learning. These are foundational concepts and being conversant in them will go a long way in starting to build your data muscle. If you can master that, you’ll be proficient enough to understand a bulk of the work likely being done around you in a consumer web company. The meat of the book is chapters 3, 4, 5 and 6 titled “Linear Regression,” “Classification,” “Resampling Methods,” and “Linear Model Selection and Regularization.” If you’re short on time and read nothing else, be sure to work your way through chapter 6. It introduces some statistical concepts as well as the intuition behind the Bias-Variance trade-off. The book starts by motivating Statistical Learning as a field. This book stands on its own, but what I got from it was multiplied by the generous time given by patient people around me. I’ve built some really strong friendships (and had a lot of fun) by pulling people into a room and asking them to help me prove an assertion in the book or to understand the intuition of a concept. I have the fortune to work with incredibly smart people who are at the top of their field in Machine Learning.
![used an introduction to statistical learning used an introduction to statistical learning](https://elitedatascience.com/wp-content/uploads/2016/10/Intro-to-Statistical-Machine-Learning.png)
![used an introduction to statistical learning used an introduction to statistical learning](https://www.bijenpatel.com/content/images/size/w2000/2020/08/intro-statistical-learning-2.png)
I read the book cover to cover stopping occasionally for a week long digression into a topic only partially covered. If you’re an engineer aspiring to make your first steps into Machine Learning I recommend you start here and find a couple of mentors. It's an alternative to the deeper coverage of the same materials given in Elements of Statistical Learning by some of the same authors. ISLR is written for practitioners giving enough theory to have substance, while omitting enough to keep you from drowning. These books felt like a depth first search of the topic. I’d explored similar books in the past, but I’d always stopped after slogging through them overwhelmed by the density of the chapters. Reading it, at this point in my career, reminded me of my first year out of undergrad sneaking into work at Ford Motor Company an hour early to borrow from the company library to discover books like The Pragmatic Programmer, Clean Code, Effective Java, Test Driven Development by Example, and others.Īfter a year on a team building recommender systems for LinkedIn Groups and Learning products, with ISLR I finally found a book that structured the many things I’ve been learning the hard way and at the right level of depth. There are a couple of books in my library that have become part of the canon of my engineering philosophy. A few of it’s chapters were being covered in a reading group at work. I stumbled onto Introduction to Statistical Learning (ISLR). I set a goal in 2016 to be able to retire that phrase.
![used an introduction to statistical learning used an introduction to statistical learning](https://static.javatpoint.com/tutorial/machine-learning/images/introduction-to-machine-learning.png)
Early in my tenure on the team I would introduce myself as a systems engineer with data aspirations. I understood the rough concepts, but much of it was still a black box. I was a novice in the data space when I joined the relevance team.
#USED AN INTRODUCTION TO STATISTICAL LEARNING HOW TO#
I was eager to not just deliver recommendations to members, but to learn how to generate them as well. I was at the boundary of the application and our relevance team who was responsible for the quality of recommendations. A lot of what I did was build systems that could ingest news recommendations and reliably deliver them to members.
#USED AN INTRODUCTION TO STATISTICAL LEARNING OFFLINE#
At the time, I was an applications engineer on our news product building mid-tier web services for mobile applications as well as offline workflows for email distribution. A little over a year ago, I began a career transition of sorts.