Happy new year! We usually send our newsletter a little later in the month, but we have a great webinar coming up, a speaking spot at TWIMLcon, and a new product to announce, so we’re feeling a little impatient. As always, our highlights first, followed by the best machine learning stuff we’ve read recently.
WHAT WE’VE BEEN DOING
Introducing the Alectio Labeling Marketplace
We’ve partnered with some of the best labeling shops in the world and are thrilled to announce our new Data LabeLabeling Marketplace.
Instead of locking yourself into long-term contracts and working on your providers’ schedule, our marketplace gives you full control. Choose the right labeler for every individual project, leverage our autolabeling solution to reduce your costs, and get better labels faster. Here’s a whitepaper with some additional details as well.
We hope you’ll check it out!
5 ways to make your ML projects actually profitable
From hiring the best team (in the best order) to streamlining your data practices to rethinking machine learning operations, there are plenty of smart ways to turn your ML initiatives ROI positive. And we’d love it if you joined our founder & CEO Jennifer Prendki for a free webinar later this month to find out how.
How active learning unlocks aerial imagery
Some fascinating research that highlights how important active learning querying strategies are for reducing the labeling costs for aerial imagery projects.
TWIMLcon talk on data prep operations
This Week in Machine Learning (a.k.a. TWIML) has long been one of our favorite podcasts. This year, Sam and company are throwing a virtual convention and we’re really proud to be speaking on the first day. You can click here for a full list of sessions.
WHAT WE’VE ENJOYED
An AI that creates images from text
DALL-E is a clever ML project where a model creates images described by text. The results are quite impressive and, at the very least, we now want an avocado chair for the office.
An AI that knows how to use real world objects
Speaking of novel machine learning projects, here’s one that focuses on models and their manipulation of real world objects.
A thoughtful interview about democratizing AI
Really enjoyed this Frank Pasquale interview about the new laws of robotics and creating a more ethical framework for AI going forward.
Responsible sourcing across the data supply line
The folks at Partnership on AI are putting together guidelines for responsible sourcing and it’s a laudable effort. Ethical, fair treatment of labelers is something we believe in here at Alectio (it’s something we prioritized when choosing partners for our marketplace, in fact) and we’re excited to dig into their recommendations in the future. You can also get involved via the link above.
That’s it for now! Thanks for reading and feel free to reply to this email or send us a note at email@example.com if you have any questions about, well, anything. Have a great January!