Let’s Talk About Math (for Data Scientists)

Math tends to elicit strong emotions among professionals with data-centric careers. Some people want to become data scientists because they love math. Others need to overcome deep-rooted anxiety around this subject before they find their footing in the profession; even then, they might aim to use their math knowledge as rarely as possible.

Regardless of your personal experience, we hope you enjoy this collection of articles that explore math’s beauty (and complexity) with passion and patience. They go from beginner-friendly to more advanced topics, so you’ll find something to sink your teeth into no matter how much (or how little) you remember from your last boot camp, university course, or—gasp!—high school class.

• An accessible introduction to logistic regression. Statistics is the subfield of mathematics that data scientists encounter most frequently, and within it, logistic regression is one of the key concepts they need to master. If you’re at the earlier stages of your learning journey, don’t miss Shreya Rao‘s wonderfully illustrated back-to-basics guide.
• [Perplexed by kernels? Not anymore](http://beautiful mathematical concepts that are used in machine learning and statistics with different forms). Kernel functions, says Shubham Panchal, are "beautiful mathematical concepts that are used in machine learning and statistics with different forms." Their versatility can make them seem confusing, but Shubham’s explanations will help you clearly see how to use them in ML applications.

• The challenge of choosing the right path. If you’re dealing with a tough mathematical-optimization problem, knowing what kind of approach is the most contextually appropriate can be difficult. Hennie de Harder‘s explainer walks us through the process of deciding between exact algorithms (think linear or mixed-integer programming) and heuristics (like genetic algorithms and particle swarm optimization).
• Unpack the intricacies of a classic probability puzzle. Naman Agrawal invites us to follow along as he sets out to solve the coupon collector’s problem, which can "challenge our understanding of the world around us." Along the way, Naman also discusses the problem’s complexity and various implications for fields as far and wide as computer science and economics.

We have a few more reading recommendations for you this week – hopefully you’re not quite mathed-out yet! Enjoy:

• Before we really leave math behind, check out Bex T.‘s detailed tutorial on creating stunning fractal art in Python.
• Yennie Jun took a thorough and unflinching look at the data behind the alarming trend of book bans in the U.S.
• Machine learning meets fashion in Federico Bianchi‘s introduction to FashionCLIP: a new domain-specific vision and language model.
• With March Madness entering its (almost) final stretch, now’s as good a time as any to read Giovanni Malloy‘s explainer on how college basketball’s NET rankings work.
• If your education or career didn’t follow a typical path, a data science job is still within reach—Madison Hunter explains how to make the most of a less-traditional background.

Thank you for your time and your support this week! If you enjoy the work we publish (and want to access all of it), consider becoming a Medium member.

Until the next Variable,

TDS Editors

Share This Article

• Share on Facebook
• Share on LinkedIn
• Share on X