Mini-Workshop: Positional Games

This mini-workshop focused on Positional Games and related fields. Positional Games Theory is a branch of Combinatorics whose main aim is to systematically develop an extensive mathematical basis for a variety of two-player games of perfect information and without chance moves, usually played on discrete objects. These include popular recreational games such as Tic-Tac-Toe and Hex as well as purely abstract games played on graphs and hypergraphs. Though a close relative of the classical Game Theory of von Neumann and of Nim-like games, popularized by Conway and others, Positional Games are quite different and are more of a combinatorial nature. The subject is strongly related to several other branches of Combinatorics like Ramsey Theory, Extremal Graph and Set Theory, and the Probabilistic Method. It has also proven to be instrumental in deriving central results in Theoretical Computer Science, in particular in derandomization and algorithmization of important probabilistic tools. Despite being a relatively young topic, there are already three textbooks dedicated to Positional Games as well as one invited talk at the International Congress of Mathematicians. During this mini-workshop, several new exciting developments in the field were presented and discussed. We have also made some progress towards solving various open problems in Positional Games Theory and related areas. Mathematics Subject Classification (2010): 91A24 (Positional games), 05C57 (Games on graphs), 91A43 (Games involving graphs), 91A46 ()Combinatorial games), 05C65 (Hypergraphs), 05C80 (Random graphs), 05D40 (Probabilistic methods), 05D10 (Ramsey theory). 2716 Oberwolfach Report 44/2018 Introduction by the Organizers Positional games involve two players who alternately occupy the free elements of a given set V , which we call the board of the game. The focus of their attention is a given family F = {E1, . . . , Ek} ⊆ 2 of subsets of V , usually referred to as the winning sets. In the general version there are two additional parameters – positive integers p and q, where the first player claims p free board elements per turn and the second player responds by claiming q free elements (in the basic version we have p = q = 1, the so-called unbiased game). It remains to specify who wins the game, each such specification leading to a standard type of positional games. There are several standard types of Positional Games. The most frequently played is probably the so-called strong game, where both players compete to be the first to claim a winning set. Both Tic-Tac-Toe and 5-in-a-row are of this type. Tools such as strategy stealing and Ramsey-type statements are of utmost value here. Strong games are well-known to be hard to analyze. Nevertheless, various interesting results on strong games were obtained recently. A close relative is the Maker-Breaker game, where the first player, called Maker, wins if he fully claims a winning set by the end of the game, while the second player, called Breaker, aims to prevent Maker from fulfilling his goal. For example, Hex can be cast into this framework. In Avoider-Enforcer games, Avoider loses if he claims a winning set, or, in other words, in order to win he has to avoid claiming a winning set to the end of the game. Much of the ground work on Avoider-Enforcer games was layed down by the organizers in three papers. Exciting progress on some of the questions that were raised in those papers was achieved recently. In recent years, various other types of positional games have attracted growing attention. For example, in Waiter-Client and Client-Waiter games, the first player, called Waiter, offers the second player, called Client, p + q board elements. Client then chooses p of these elements which he claims and the remaining q elements are claimed by Waiter. Client wins (respectively, loses) the Client-Waiter (respectively, Waiter-Client) game if he fully claims a winning set by the end of the game. Typical general results in Positional Games include winning criteria for one of the players, in some cases also supplying an efficient winning strategy for that player. The proofs utilize an array of various combinatorial arguments (Ramsey Theory, Extremal Graph and Set Theory, etc.); sometimes – perhaps somewhat surprisingly – probabilistic strategies are used to analyze completely deterministic games of perfect information. This connection was first indicated by Paul Erdős. Subsequently, it was discussed in detail and masterfully implemented by József Beck. Recent developments in the field have affirmed the crucial role of probabilistic arguments in positional games. The mini-workshop on Positional games was attended by 17 people, arriving from various geographic regions (namely, England, Germany, Israel, the Netherlands, Poland, Serbia, Switzerland, and the United States). The participants had different backgrounds in Positional Games Theory (though all of them had a considerable level of familiarity with the field) and different levels of research experience ranging from M. Sc. students to Full Professors. Mini-Workshop: Positional Games 2717 In the first few days, most participants gave research talks (13 in total) which presented many interesting new developments in Positional Games and related fields (in particular, in Ramsey Theory and Random Graphs). The talks also included many open problems which indicated new research directions to be explored. In the evening of the first day of the mini-workshop we held an open problems session in which participants offered many “good” open problems (some were good problems in the sense that solving them is likely to have an impact on the field, others were good in the sense that it seemed plausible one could make some progress towards solving them in a week, some were perhaps good in both ways). After this session, the participants were divided into four groups. Starting on the second day, these groups have engaged in focused open problem solving activities. This continued until the end of the week and will hopefully continue (in one way or another) for a much longer period. We have tried to form the groups in a way which will foster new and lasting collaborations between researchers with different levels of experience. All groups have reported some progress during the week and we expect several publications to result from this mini-workshop. All in all, we believe the mini-workshop was a great success. Acknowledgement: The MFO and the workshop organizers would like to thank the National Science Foundation for supporting the participation of junior researchers in the workshop by the grant DMS-1641185, “US Junior Oberwolfach Fellows”. Mini-Workshop: Positional Games 2719 Workshop: Mini-Workshop: Positional Games