ISSUE 397  ·   July 26, 2022

🔥🔥🔥

Do data-driven companies actually win?

"...many of us bet our careers on data being necessary for corporate survival. But if we divorce ourselves from our employer and our career ambitions, and truly put these beliefs to the test ... do we still stand by them?" Great post!
Benn Stancil

Tutorials, Projects & Opinions

The Open Loop of ML

Great series of posts that explores the mismatch between machine learning systems with high "accuracy" and real-world results. Parts one and two describe the issue and Part three suggests a new measure of performance that's not based on model "accuracy."
Devesh Rajadhyax

Short Chronology Of Deep Learning For Tabular Data

Tabular data is essentially the opposite of unstructured data and unstructured data is where deep learning shines. Even so, researchers play with special-purpose deep learning methods for tabular datasets. This is a nice summary of recent research and what, ultimately, you should do instead. Includes highlights and links for key papers.
Sebastian Raschka

Building Human Values into Recommender Systems

Recommender systems select, filter, and personalize content across many of the worlds largest platforms. What human values should recommender systems support? How should their outcomes be measured? What designs would achieve this? And what policies would incentivize it? This interdisciplinary paper explores these questions across domains with an eye on current industry practices.
arXiv | Jonathan Stray, et al.

Invest in your professional goals with Coursera Plus

Level-up your skills with unlimited access to 7,000+ world-class courses, hands-on projects, and job-ready certificate programs. Courses include the popular Learning How to Learn, Applied Data Science, Mathematics for Machine Learning, Data Analytics, Python 3, and many more. One low-cost subscription is all-inclusive.
// partner link

Resources

Guide to Sports Analytics:  ⚽ 🏀 ⚾ 🏈 🏒 ⛳ 

Awesome collection of sports analytics resources covering a wide range of sports. Includes tutorials, research papers, packages, cheatsheets, free books, accounts to follow and more. Covers Python and R. Be sure to check the tabs at the bottom of the page.
Google Doc | Dom Samangy

Practical Deep Learning for Coders 2022

This popular course from fast.ai has been completely rewritten for 2022. In this new release, there's a bigger focus on interactive explorations and it covers a broader mix of libraries and architectures. It's a free online course with a downloadable book, lecture videos, and even a learning community. 
fast.ai | Jeremy Howard

Data Visualization

How Florence Nightingale Changed Data Viz Forever

Many people know her name but most don't know that Florence Nightingale was a key influence in the emerging field of data visualization — in the 1850's! At the time, people weren't interested in tables of numbers and her story was life and death. Here's how it went, including prints of her original hand-drawn graphics.
Scientific American | RJ Andrews

Outlier

Tensor Puzzles 🧩

Like chess puzzles, these Tensor Mini Puzzles aren't meant to simulate the complexity of a real program, but to practice in a simplified environment. Each puzzle asks you to reimplement one function in the NumPy standard library without magic.
GitHub | Sasha Rush

How did you like this issue of Data Elixir?

👎    1     2     3     4     5   👍

Was this email forwarded to you?  Sign up here >>