Two recent collaborations between mathematicians and DeepMind demonstrate AI's potential to help researchers generate new mathematical conjectures.

Math and Science News from Quanta Magazine View this email in your browser MACHINE LEARNING | ALL TOPICS   Machine Learning Becomes a Mathematical Collaborator By KELSEY HOUSTON-EDWARDS Two recent collaborations between mathematicians and DeepMind demonstrate the potential of machine learning to help researchers generate new mathematical conjectures. Read the article PARTICLE PHYSICS   The Mysterious Forces Inside the Nucleus Grow a Little Less Strange By CHARLIE WOOD The strong force holds protons and neutrons together, but the theory behind it is largely inscrutable. Two new approaches show how it works. Read the article Related:  What Goes On in a Proton? Quark Math Still Conflicts With Experiments. by Charlie Wood (2020) SEX   A Billion Years Before Sex, Ancient Cells Were Equipped for It  By JAKE BUEHLER Molecular detective work is zeroing in on the origins of sexual reproduction. The protein tools for cell mergers seem to have long predated sex — so what were they doing? Read the article Related: Why Sex? Biologists Find New Explanations. by Christie Wilcox (2020) NEURAL NETWORKS   AI Overcomes Stumbling Block on Brain-Inspired Hardware By ALLISON WHITTEN The biggest neural networks consume nearly as much power as five cars do over their lifetimes. New work may help bring the brain's energy-efficient computing to AI. Read the blog Related: To Be Energy-Efficient, Brains Predict Their Perceptions by Anil Ananthaswamy (2021) QUANTA SCIENCE PODCAST   Mathematicians Outwit Hidden Number Conspiracy Podcast hosted by SUSAN VALOT; Story by JORDANA CEPELEWICZ Decades ago, a mathematician posed a warmup problem for some of the most difficult questions about prime numbers. It turned out to be just as difficult to solve, until now. Listen to the podcast Read the article Around the Web Narrowing a Massive Mystery A new measurement of the neutrino's mass places the tightest bound yet on its upper limit, at just 0.8 eV. Davide Castelvecchi reports for Nature on the newest result from the KATRIN collaboration. Figuring out the mass of the neutrino could help solve big mysteries about the Standard Model of particle physics. The KATRIN experiment is trying to do just that, Marcus Woo wrote for Quanta in 2019. Putting the Art in Artificial Intelligence Researchers are using AI for creative tasks like painting and composing original music, reports Richard Moss for Science News. Some think that injecting creativity into AI algorithms will be the key steppingstone to building truly intelligent machines. For an algorithm's solutions to be creative and open-ended, the problems must be open-ended, too. Matthew Hutson wrote about it for Quanta in 2019. Follow Quanta Simons Foundation 160 5th Avenue, 7th Floor New York, NY 10010 Unsubscribe | About Quanta Privacy Policy | Terms | Contact Us   Copyright © 2022 Quanta Magazine, an editorially independent division of the Simons Foundation