Machine learning helps researchers capture sharper image of black hole

Researchers have used a machine learning algorithm to produce a sharper image of a black hole located at the center of the galaxy Messier 87, which is over 53 million light years away from Earth.

The original image was captured in 2017 by the Event Horizon Telescope (EHT), but the researchers used a novel algorithm called PRIMO to reconstruct a more accurate image. The new image shows a thinner ring of light and matter surrounding the black hole's center, consistent with Albert Einstein's predictions.

PRIMO is a dictionary-learning-based algorithm that was trained on simulations of over 30,000 black holes. It learns to "recover high-fidelity images even in the presence of sparse coverage" by using data based on what we know about the universe's physical laws, and black holes in particular.

The algorithm analyzes the correlations between different parts of the image to produce a better-looking and more accurate shot from the raw data captured in 2017.

The research team, led by astrophysicist Lia Medeiros of the Institute for Advanced Study in Princeton, New Jersey, says that the new image is consistent with Albert Einstein's predictions. However, they expect further research in machine learning and telescope hardware to lead to additional revisions.

"In 20 years, the image might not be the image I’m showing you today," said Medeiros. "It might be even better.