A new scientific approach uses a model inspired from Google's algorithm to determine the degree of stability of the planetary systems. The researchers employed a machine learning technique in order to create a mechanism that is 1,000 times faster than traditional methods when it comes to predicting stability.
The study, published Nov. 23 in the journal Astrophysical Journal Letters, was carried out at the University of Toronto Scarborough by a team of scientists, and it employed a novel instrument in the analysis of planetary stability.
Using Machine Learning In Analyzing Planetary Stability
Machine learning is an artificial-intelligence based technique which allows a computer to learn without being programed to carry out a particular task or set of tasks. Its major advantage lies in the dynamism of the instrument when in contact with new data.
The processing is done more rapidly and more efficiently through this method. Machine learning has also become widely used in the 21st century in various industries, from preventing fraud and detecting email spam in Google to Netflix movie recommendations.
"Machine learning offers a powerful way to tackle a problem in astrophysics, and that's predicting whether planetary systems are stable," noted Dan Tamayo, lead author of the research.
As part of the research, the team of scientists developed a series of workshops at the university, in an attempt to detail the way machine learning can be employed in analyzing planetary stability.
Previous Techniques - Slower And Less Efficient
"What's encouraging is that our findings tell us that investing weeks of computation to train machine learning models is worth it because not only is this tool accurate, it also works much faster," adds Tamayo.
Previous approaches have all struggled with the interpretation and analysis of such a massive amount of data, which makes it nearly impossible to see the bigger picture. While progress was made within the instruments that were used for this process, the new study revolutionizes the practice through the nearly endless capacity of analysis and synthesis that machine learning offers.
The importance of understanding whether planetary systems are stable or not lies in the correlation that can be made between this particular behavior and their formation. This type of analysis can also provide insight on exoplanets that are currently unobservable through our existing means.
While a series of methods employed in the detection of exoplanets currently exists, they are not always reliable, as they don't offer significant details about their structure and characteristics, such as the mass or the degree to which their orbit is elliptical, which are essential in understanding stability.
