Hey everyone. I hope you are well. In this issue, we dive into an ML research essay on the social dilemma, some exciting deals to make your AI-holiday shopping even better, the misconceptions of AI and the challenges of natural language generation (NLG), an updated chart of all significant neural networks, and the ML research paper highlight of the month.
This issue is brought to you thanks to our friends at Amazon Science:
Check out the Amazon Science website to learn more about the company’s unique approach to customer-obsessed science, and how it helps attract some of the brightest minds in artificial intelligence, machine learning, and related fields. Find blog posts and research papers from Amazon scientists and academics, including which conferences they’ll be attending, and how to collaborate. View available jobs.
Before we get started, I wanted to let you know that we have decided to publish the latest machine learning research every weekday after 8 PM ET. All based on your opinion, it seems that you are even hungrier for more — research in AI, CV, NLP, and others. So stay tuned; we’ll keep you updated with what we decide on doing to keep you up to date.
All right, so let’s get to it.
An Essay by ML Researchers on “The Social Dilemma”
Deja Vu much? Researchers at Carnegie Mellon argue on the blog “When Curation Becomes Creation: Algorithms, microcontent, and the vanishing distinction between platforms and creators” the challenges of implementing the right policies and regulations that balance ethics, the economy, individual rights, and proprietary data. The thorough essay showcases the gray areas between every piece of content published and distributed on the internet and what we can do about it.
AI-Holiday Deals
Ho ho ho! I hope I don’t bore you too much with these. But! In case you are looking for a new AI rig, we just updated our shopping recommendations for deep learning laptops or AI workstations. So please take a look, and as always, all feedback is welcome — if you do get one (and have browser cookies enabled), you’ll be supporting us, and we genuinely appreciate it.
ML Research Paper Highlight of the Month
Wolfgang Konen and Samineh Bagheri from the University of Applied Sciences in Germany published a cool paper called “Final Adaptation Reinforcement Learning for N-Player Games,” which concentrates n-tuple based reinforcement learning algorithms for games, specifically new ones tackling TD-, SARSA-, and Q-learning. All reproducible and open-source code for the paper is available on Github.