Machine Learning May Help Find Dark Matter

What is required to discover the dark material-the invisible and developed matter that may constitute five sixth of the matter in the universe? The dark matter around us must be, which has small effects on the regular material, but the searches have appeared empty so far. But a new study indicates this A The strategy that uses machine learning can help quantum sensors to search for them. This is amazing Other sensitive sensors may also have other applications, such as GPS free navigationDiscover underground warehouses, discovery Attractive ripples In space time after the big explosion.

We know that the dark matter exists due to its gravitational effect on the movement of galaxies. But we do not know what is made, or how it can interact with the daily particles that form you or. Although scientists dreamed of dozens of models to form a potential dark matter and accurate properties, most of these proposals predict small effects on natural particles. One of the potential to detect these small reactions is quantum sensors. The quantum effects are vulnerable to external interference, and quantum sensors benefit from this fragility in order to respond to the lowest disorders in the environment – such as the interaction between the dark matter and the natural material.

The new study focused on the atom’s overlap measures, a type of quantum sensors that depend on an effect known as the overlap, where one atom can be found mainly in two or more places at the same time. The sensors contain the cases of the cat such cats, where the corn walks in separate paths and then reassembled. Because of the duplication of the wave of particles-the phenomenon of quantum where molecules can behave like waves, and vice versa-these atoms overlap with each other, with the peaks of their waves, suppression, or another increase. The examination of this overlap can reveal the extent of the a little different movements that took place on different trips.

One of the approaches that scientists use to enhance these interference standards depends on the laser pulses that these sensors use to divide the corn waves from each other and reflect them in each other. These laser pulses take the place of traditional mirrors in visual interference. Every time the waves of atoms bounce back to one of these fast mirrors, “The sign that we are looking for can enlarge, such as how light signals can be enlarged when they bounce in a lumen lined on the mirror,” says Timothy Kovaci.

However, the number of times you can face the atom wave such repercussions depends on the quality of the corn mirror, and “it is difficult to make a good corn mirror,” says Kovaci.

Now, in the new study, Kovaci and his colleagues have revealed a strategy to enhance the number of possible repercussions of atomic mirrors. Using machine learning, instead of reflecting corn waves of a sequence of approximately 10 laser pulses at most, the new approach allows a sequence of approximately 500 pulses.

The most sensitive interference devices for corn

The new strategy “does not insist on making the perfect corn mirror.” “Instead, looking for a way to improve the collective voltage of many different mirrors of the atom, which compensates for the defects of each individual atom mirror.” The result is the improvement of 50 times in the performance of the overlap in corn in laboratory tests.

“When we started this work, I didn’t really imagine it would be possible to get this degree of improvement,” says Kovaci. “It is always good when there are fun surprises.”

The researchers now hope to implement their new style through “the first major search campaigns for dark matter with ATOM scales, which are currently being constructed,” says Kovacachi. “We expect the first searches on the Internet within three to five years or so. We hope, in conjunction with better corn optics, to enhance its sensitivity through multiple orders in terms of size on what can interfere in the atom now.”

Kovaci says that the most accurate atomal overlap devices may also have other applications, such as GPS. Satellite links that help enable global satellite systems for navigation in global navigation underwater or underground, and where they work, are vulnerable to confusion, deception, and weather. The motion sensor can help as a basis for the navigation system with self -deficiency that does not depend on any external signals.

Kovachy adds that overlapping overlapping devices can also help measure the strength of the Earth’s gravitational field, which can vary across the surface of the planet depending on the amount of mass that focuses under it. Possible applications of gravitational sensors include seeing the underground structures, discovering ground natural resources, discovering underground archaeological sites, monitoring volcanic activity and groundwater flows.

“I didn’t really imagine it would be possible to get this degree of improvement.” – Timthi Kovaci, Northwestern University

Wide -scale interference standards of 100 meters or even large (comparison, the length of the standard overlap of corn 1 or 2 meters) is under development, even one day may help discover ripples in space known as gravitational waves. Scientists discovered the first direct evidence of these waves using the LIGO Observatory in 2015, most likely when two black openings collide. Theoretical intervention scales can theoretically discover gravitational waves from various events, such as the mysterious proposed era known as inflation, when the universe underwent the Titanic growth mutation after moments of the big explosion, says Kovashi.

Future research should achieve this new technology with different types of corn interference standards. In the new study, scientists have tried a device that depends on sterintium atoms, but “Rubidium atoms are definitely the backbone of measuring the overlap in the atom,” says Kovaci.

The scientists detailed their results December 11 in the magazine Physical review messages.