— In the News —
Ultimately, success with AI requires a lot of upfront, manual labor and China has a large and relatively inexpensive labor force to work with. Here's how China expects to become the world leader in artificial intelligence by 2030.
— Profiles —
Great interview with Angela Bassa about her approach to building and growing data science teams, career development, and things to watch for along the way. Angela is the Director of Data Science at iRobot and this interview is rich with insights and inspiration.
— Sponsored Link —
Data science and analytics positions are taking the job market by storm. A master’s
degree or graduate-level certificate from UW–Madison can jump start your data career. We have 13 online, one-year, full-time, or part-time data programs to fit your life and help you succeed, including GIS, Data Analytics and Actuarial Science. Explore more today.
— Tools and Techniques —
Great article about the rapidly growing field of people analytics. Here's what all the hype is about, what many teams get wrong, key ways to consider the inter-connections at your organization, and what to think about when gathering the data.
In her latest post, Emily Robinson explores some packages and functions that are useful for exploratory analysis but aren’t as well known as the core parts of ggplot2 and dplyr.
This step-by-step notebook tutorial is an easy introduction to privacy preserving, decentralized deep learning. Here's how to do things like run ML spellcheck on encrypted email.
Nice explainer by Sebastian Raschka on how decomposing bias-variance can be useful for describing the performance of a model. This goes along with a new bias-variance decomposition function that he added to his mlxtend library.
— Resources —
Amazon's program for teaching machine learning to its own engineers is now being made freely available to everyone via AWS. The program has more than 30 online courses with distinct learning paths for developers, data scientists, data platform engineers, and business professionals.
— Data Viz —
"Literacy" is used sometimes to help guide the design of data visualizations but that word functions sometimes as a way to shift blame for bad charts to "illiterate" audiences. Here's a thoughtful perspective on making sure our visualizations are understood.