Destiny of science modeled and explained in new study
Jul 2022, phys.org
Researchers modeled the evolution of convergence by analyzing millions of scientific works using machine learning etc. to identify several stages in the evolution of science, each characterized by a different form of convergence:
- Polymathic Convergence - early science to the Renaissance period, polymaths Aristotle and da Vinci, knowledge integration within the minds of singular scholars
- Disciplinary Divergence - theories developed within specific disciplines turned into generalized templates with broader applications, "convergence through divergence", Darwin's theory of evolution in biology
- Multi-Disciplinary Team Convergence - mid-20th century, experts from different disciplines work together, knowledge integration takes place across teams of scientists with diverse expertise, the Manhattan Project.
- Polymathic Team Convergence - knowledge integration both within and across scholars, a mix of individual polymathic and multi-disciplinary team convergence, recent brain science research.
- Cyborg Team Convergence - polymathic scientists collaborate with artificial intelligence in mixed human-machine teams.
Note to Self: We are already at stage 5, because machine learning networks, although designed by scientists, are discovering things that we never knew, and sometimes things that we didn't even ask about, such as the Foodome project at Barabási Labs that discovered an unknown formula in the nutrient profiles of food, or the UC Berkeley crew that discovered a formula related to exoplanets hidden in the math of general relativity, or the DeepMind project that revealed a new theorem about knots (yes knots, trust me).
via University of Houston: From Polymaths to Cyborgs - Convergence Is Relentless. Ioannis Pavlidis, Ergun Akleman, Alexander M Petersen. American Scientist, July 2022.
Image credit: Black Hole Merger Gravitational Waves - NASA C Henze - 2022
When I see the picture above, which somes to us via NASA, I think to myself "Arthur Dove meets Charles Burchfield", and now that we've figured out how to make robots metabolize and regurgitate our cultural artifacts back to us in visual form, we can see what that looks like:
Post Script: Good News
Government-funded scientific research reflects public interest, study finds
Jul 2022, phys.org
Jones and Wang also noted that the study emphasizes the value of using the large datasets available today to study other high-level questions about science as a public good in ways that were not previously possible.
via Northwestern University: Yian Yin et al, Public use and public funding of science, Nature Human Behaviour (2022). DOI: 10.1038/s41562-022-01397-5
No comments:
Post a Comment