Artificial intelligence, or simply AI, is starting to rapidly take over the world of technology. Whether it is an assistant like Siri or Alexa, or a self-driving car produced by Tesla, AI is changing our everyday lives. Although today’s AI is mostly used for face recognition or internet research, in the years to come it will be designed to outperform humans at every cognitive task there is.
The most recent examples of AI “beating” human intelligence are in games like chess or poker. For instance, playing poker is dealing with loads of imperfect information which makes the game complicated. This is the main reason why the AI departments of many universities are researching poker. The University of Alberta, the Carnegie Mellon University and the University of Auckland are the driving forces of the research.
Libratus, researched by Carnegie Mellon University, won a game of Texas Hold’em against human professional card players in 2017.
It is considered the biggest achievement of an AI program.
Libratus has three modules, each performing a different task. The first one learns the poker rules and is trying to find strategies that would work for every situation. The second one is the decision maker for each action during a hand. And the last one makes consistent updates, records and folds new information from each move into the system. Compared to other AI’s, Libratus doesn’t have a deep neural network; it is learning the game rules from scratch and is strategy independent from human play.
The technology market, government and defense, banking, healthcare, cybersecurity, etc., are some of the sectors that are starting to adopt the AI technology source. The world around us is changing and we can say that there will be winners and losers as AI technology “marches” forward with big steps. The main challenge will be discovering which aspect would benefit our society the most.