A Welcome from the President, Jonathan Schaeffer

Welcome to the site of the International Computer Games Association (ICGA). Our mission is to advance the state of the art in artificial intelligence technologies as demonstrated using games.

Why games? They’re fun! Since the dawn of the computing age, games have been extensively used as experimental test-beds for demonstrating the effectiveness of ideas in artificial intelligence (AI). The work is mainly in the areas of search algorithms, knowledge representation, machine learning, parallel computing, and explainable AI. Games are a nice domain to work with. The playing area is well defined, the rules are fixed, they are easy to test, and there are no real-world consequences if the program plays badly (other than a hurt ego).

Why AI? Because AI is the future. Yes, AI is popular in the media these days but it wasn’t always that way. Our community understood the importance of AI and carved out a niche where we could advance the state of the art in a fun, educational, and beneficial way. Over the years, games-related research has resulted in many contributions to AI knowledge. For example, the ideas of iterative deepening and transposition tables were pioneered in chess programs, yet variants of these enhancements are essential to many popular search algorithms today. As a more up-to-date example, reinforcement learning is a powerful tool today, but it wasn’t until the mid-1990s that the backgammon program TDGammon was used to demonstrate the potential for this algorithm. As they say, the rest is history. And now, the combination of reinforcement learning with so-called deep learning has set the entire field of AI buzzing with excitement.

A further benefit of our work is that the public understands the combination of games and AI. The AI literature is full of obscure/artificial/uninteresting AI test domains that are valuable to scientists but incomprehensible to the general public. In contrast, the combination of games and AI is easily understood by most people. The specter of a solitary man matching wits with an electronic computing monster is something we can all relate to, a David-and-Goliath story. All the major game-AI successes have attracted immense media attention, such as for checkers, chess, Go, and poker. Two of the biggest AI media stories in history are to do with games: the 1996/1997 chess matches between Garry Kasparov and IBM’s Deep Blue; and the 2016 Go match between Lee Sedol and DeepMind’s AlphaGo.

The first computer-only competition was in chess. In 1970 the first ACM Computer Chess Championship was held. Since then – 50 years and counting – we are still holding computer chess tournaments! Because of the great interest in computer chess in the formative days of computing, the International Computer Chess Association (ICCA) was created. Briefly, its mission was to encourage research into game AI and support competitions to evaluate game AI. After the 1997 Deep Blue match, the ICCA morphed into the ICGA – substituting “games” for “chess”, appealing to the interests of a broader community. To this day, the ICGA is true to its goals.

I am delighted to be contributing to the ICGA as its President. I joined the ICCA in 1983 and have been a continuous member since then. The ICCA/ICGA has been a tremendous support for my academic career, providing me venues to publish my research (ICCA/ICGA Journal; Advances in Computer Chess/Games conferences) and to assess the results (computer chess tournaments; Computer Olympiads). It is important that I give back to the community that has been so good to me.

If you have any interest in programming AI for games, consider joining the ICGA. It does not matter whether you are an academic, student, researcher, hobbyist, whatever. Here is a chance to write programs, compete on the world stage (maybe even for a World Championship), and interact with others who share your passion.

I am taking over from David Levy as President. He has led the ICCA/ICGA for 26 of the organization’s 45 years. During his time, he achieved many successes. Clearly I have work to do! All of us are greatly appreciative of his work on behalf of our membership.

We are always interested in hearing from the membership – past, present, and future. The best way to reach me is by email: jonathan@ualberta.ca. I look forward to growing the ICGA and adding value for its members.

Comments are closed.