2025 - 2026 Colloquium

We have all heard the saying “If you cannot measure it, you cannot make it,” for optical fabrication, but it applies to most things where precision is required. The past decades have seen fantastic advancements in optical systems from astronomical telescopes that see to the edge of the universe to virtual reality systems that enable people to experience worlds that never exist. The new optical components and systems have required new ways to measure them.

Recent strides in machine learning have shown that computation can be performed by practically any controllable physical system that responds to physical stimuli encoding data [1]. This perspective opens new frontiers for computational approaches using Physical Neural Networks (PNNs) [2, 3, 4] and provides a framework to deepen our understanding of their biological counterparts—neural circuits in living organisms.

One of NASA’s primary science goals is to directly image and characterize the atmospheres of Earth-like planets orbiting nearby stars. The observations are extremely challenging because the planets appear adjacent to their parent stars, near telescope diffraction limits, and their reflected light is 10 billion times fainter than the star. An elegant solution to this problem, first proposed by Lyman Spitzer in 1962, is to employ a starshade, a flower-shaped diffraction screen positioned far in front of a telescope so that it shadows the starlight without blocking the planet light.