Paper for this week
In line with the lots of news on human-nonhuman relationships and forms of intelligence, this paper fits the discourse.
“Building artificial intelligence (AI) that aligns with human values is an unsolved problem. Here we developed a human-in-the-loop research pipeline called Democratic AI, in which reinforcement learning is used to design a social mechanism that humans prefer by the majority.”
Koster, R., Balaguer, J., Tacchetti, A., Weinstein, A., Zhu, T., Hauser, O., … & Summerfield, C. (2022). Human-centered mechanism design with Democratic AI. arXiv preprint arXiv:2201.11441
See you next week; still not on vacation :)