💡Deciphering AI research for you
💡Deciphering AI research for you
We're bridging the gap to foster AI implementation
Last year alone, over 60,000 AI papers were published.
Breakthroughs are coming at a rapid pace. How are business and product leaders, and their data science teams, supposed to make sense of it all in order to properly experiment and implement AI? In fact, surveys show that one of the obstacles in AI implementation is the lack of understanding among managers of how the technology works.
I invited a seasoned AI Researcher Libby Kinsey to distill the most important AI and ML research which practitioners should know about at this moment.
The resulting AI Currents report is accessible to those with a strong AI acumen, as well as those who do not speak the language of research.
The report is broken into five segments:
A Spotlight on Transformers
From Text Sequences to Mathematical Sequences
Putting Models Into Production
An Exciting Application
Reinforcement Learning that Works in Real-World Environments
Each section answers why the research matters, what its implications will be in the next year, and serves as a repository of resources for anyone who wants to go deeper.
AI Currents was produced by Marija Gavrilov.
Libby, Marija and I are looking to hear your feedback, questions and certainly hope that this report helps you see new opportunities.
Very useful update. Many thanks Azeem and Libby.
Great report, especially the selection of papers grounded in real-world impact. MuZero is extremely though-provoking: can we really have the super intelligence of AlphaZero consistently applied to messy problems, such as societal ones? I am thrilled to see how, if that happens, it cascades across organisations and institutions. Although, the hardness of cracking the missing pieces is very hard to estimate, I am curious to hear if anyone here has any informed guess about it.