Swarm Intelligence Found To Accurately Predict Winners Of Kentucky Derby

By R. Siva Kumar - 12 May '16 12:38PM
Close

Artificial intelligence can be a game-changer through an amazing, new application---gambling.

The AI company Unanimous harnessed a software platform called UNU last weekend to assess and predict Kentucky Derby finalists.

The experiment  had 20 participants who used Unanimous to narrow the field of 20 horses. The software has predicted even Oscars and the Super Bowl winners, and now it predicted the winning order of the Kentucky Derby---Nyquist, Exaggerator, Gun Runner and Mohaymen. The prediction was 100 percent accurate.

Not only the winner but the exact order of the three horses behind it were predicted. The odds of guessing the Kentucky Derby Superfecta correctly are 540-1.

Still, the Superfecta was guessed correctly, winning UNU inventor Louis Rosenberg a $10,842 payout from a $20 bet and TechRepublic reporter Hope Reese $542.10 from $1.

"I placed my $1 bet on the race at the Derby on Saturday and made $542.10 - the odds of winning the superfecta [the top four finishers in order] were 540-1," he said.

The software uses a unique form of swarm intelligence that amplifies without replacing human intelligence. A group of people logged into a UNU online forum through their smartphone or computer.

They were given a question and a set of possible answers. Each participant was given a graphical magnet that could drag a puck across the screen to what they saw as the correct answer. As there is just one puck, it might fall on any one answer. Thus, the group needed to arrive at a collective answer that suited everyone. With just 60 seconds, the group could reach any verdict.

The application of swarm intelligence has been inspired  by bees, which use their type of swarm intelligence to locate a new home. UNU's algorithm taps into the collective knowledge and intuition to arrive at a unified answer.

Rosenberg hopes to use UNU even for health care and politics. "Politicians have conflicting values but not conflicting knowledge," Rosenberg says. "Forcing polarized groups into a swarm allows them to find the answer that most people are satisfied with. Our vision is to enable the power of group intelligence for everybody."

Fun Stuff

Join the Conversation

The Next Read

Real Time Analytics