Double-blind telescopic time
An unprecedented number of first-time investigators have secured viewing time on NASA’s Hubble Space Telescope in the years since the agency reviewed the application process to reduce systemic imbalances. In 2018, NASA changed the way it evaluates time-observation requests on Hubble by introducing a ‘double-blind’ system where neither applicants nor reviewers evaluating their proposals know each other’s identities. All of the agency’s other telescopes followed suit the following year. The measure was intended to reduce discrimination on the basis of gender and other factors, including inequality towards researchers who are in small research institutions or who have not received NASA scholarships before. Data from the Space Telescope Science Institute (STScI) in Baltimore, Maryland, which administers Hubble, shows that since the change was introduced, several first-time chief investigators have secured viewing time on Hubble.
This mathematician is an AI
How do mathematicians come up with new theories? Entering Nature, researchers describe a way to use artificial intelligence (AI) to help with the creative core of the process. After recognizing a possible pattern in the properties of mathematical objects, such as convex polyhedra (3D shapes with flat surfaces, straight edges, and points that all point outward), mathematicians typically go through a cycle to understand this pattern. They first calculate the properties of some simple examples and analyze the possible relationships between these properties. Then they refine the relationships. For example, they can come up with Euler’s polyhedral formula, which assumes that the number of corners (V) minus the number of edges (E) plus the number of faces (F) of a convex polyhedron is always equal to two: V – E + F = 2. They then test this suggested relationship on more complicated examples, discarding irrelevant properties, and trying to understand why the relationship holds. If it remains unclear, mathematicians consider various examples and the cycle continues. Researchers have now shown that machine learning techniques can help with the refinement step, which is usually heavily dependent on human intuition.
A diet to save the planet
More than 2 billion people are overweight or obese, mostly in the Western world, and 811 million do not get enough calories or nutrition, mostly in low- and middle-income countries. At the same time, the current industrialized food system emits about a quarter of the world’s greenhouse gas emissions and has other environmental impacts – which appear to be increasing as the world’s population increases and more people start eating as Westerners. With both problems in mind, nutritionists reviewed the literature to make a basic healthy diet composed of whole foods. The team then set environmental limits for the diet, including carbon emissions, biodiversity loss and the use of fresh water, soil, nitrogen and phosphorus. They ended up with a diverse and mostly plant-based meal plan that was meant to be nutritious and sustainable – a ‘planetary health’ diet. However, further studies showed that due to the cost, it would not be possible to switch to that diet in many regions of the world.