Big Data Analytics

Machine Learning on the Cloud and the Search for E.T.

Share this post:

IBM's Francois Luus teaching machine learning to help the public find aliens.

IBM’s Dr. Francois Luus teaches machine learning to help find aliens.

Maybe E.T. the Extra-Terrestrial would have made it home faster if cloud computing was readily available in 1982.

For those of you too young to know, E.T., the loveable alien in the blockbuster movie of the same name, was stranded on Earth after being abandoned its interstellar botanist colleagues.

Eventually, E.T. is found in a tool shed by some local kids who help him communicate back home using the ultimate long-distance hack: a speak & spell, an umbrella lined with aluminum foil and a coffee can filled with electronics.

But that was more than 30 years ago and today we have advanced technologies like machine learning and cloud computing to find aliens, right? Well, actually, yes.

Last week at the annual Centre for High Performance Computing Workshop in East London, South Africa-based IBM Research scientist Dr. Francios Luus hosted a three-hour session on deep learning computing environments and unsupervised machine learning.

Luus introduced a dozen participants to IBM Bluemix Spark, which will be used to analyze a condensed dataset originating from 6 million signal samples collected from the Allen Telescope Array, which was designed to be highly effective for simultaneous surveys undertaken for searching for extraterrestrial intelligence at centimeter wavelengths. Their goal is to use Spark and machine learning techniques to find anomalies in the data, which could lead to the discovery of extraterrestrial life.

“What better incentive is there to learn machine learning, then to possibly discover alien life?” said Luus.

Students were instructed to find search the data set to find a picture of an alien showing the peace sign

Students were instructed to search the data set to find a picture of an alien showing the peace sign

“It’s simply too much data for any one person to process, so we are making it accessible to anyone. This workshop will introduce the application of deep learning techniques in addressing this challenge by helping the participants to discover and contextually visualize outlier waterfalls, i.e. anomalies, which could indicate life. It’s incredibly exciting.”

The effort is part of a collaboration between IBM and the SETI Institute in Mountain View, California to use IBM’s data storage services and the power of IBM’s Bluemix Spark service to analyze the many terabytes of radio-telescope data they’ve acquired over the past few years.

The team has constructed a system, built with the IBM Cloud and Github to provide the raw data to the public, along with the tools to process it — check out the Github data set here.

Luus is hopeful of a big turnout from the public.

“If the SETI team missed any anomalies in the data, machine learning will find them. This isn’t man versus machine, it’s man plus machine and maybe, some aliens.”

More stories

The remarkable work of women scientists and researchers at IBM Research

During the month of March, IBM Research put the spotlight on a number of women scientists and engineers, and asked them about their professional and personal motivations, journeys and experiences as women — and particularly, as women in STEM. They represent the breadth of career experiences at IBM Research, across disciplines, geographies, ethnicities, tenures and backgrounds, who share a passion for science and tech, as well as a commitment to help all women rise to meet their aspirations.

Continue reading

Hybrid cloud for accelerating discovery workflows

Hybrid cloud could ultimately enable a new era of discovery, using the best resources available at the right times, no matter the size or complexity of the workload, to maximize performance and speed while maintaining security.

Continue reading

IBM AI helps to break down massive code to ease cloud migration

We use AI to automatically break down the overall application by representing application code as graphs. Our AI relies on Graph Representation Learning – a popular method in deep learning. Graphs are a natural representation for software and applications. We translated the application to a graph where the programs become nodes. Their relationships with other programs become edges and determine the boundary to separate the nodes of common business functionality.

Continue reading