In a lab where crystals are studied, a technician stares at a screen covered in bright spots and slowly rotates the sample, trying to get the spots to form a symmetric pattern. It's like tuning an old radio: you turn the knob until the static gives way to a clear signal. Now, a machine can turn that knob.
Scientists have created a program that learns to align crystals without any knowledge of how those spots work. The program operates by trial and error: it rotates the crystal, looks at the image, and gets a reward if the pattern becomes more symmetric. Gradually, it learns which movements lead to success and starts acting like an experienced technician.
Remarkably, the program, without knowing the theory, developed the same techniques as humans: it first makes coarse rotations, then fine adjustments. This means a machine can pick up human skill just by looking at images. Such an approach could lead to smart instruments that calibrate themselves for experiments, freeing scientists from routine work.
Crystals are aligned using diffraction patterns — patterns of spots that arise when X-rays bounce off atomic planes. These patterns are like fingerprints: they reveal how atoms are arranged inside.