Modern Tabletop Games present various interesting challenges for Multi-agent Reinforcement Learning. In this paper, we introduce PyTAG, a new framework that supports interacting with a large collection of games implemented in the Tabletop Games framework. In this work we highlight the challenges tabletop games provide, from a game-playing agent perspective, along with the opportunities they provide for future research. Additionally, we highlight the technical challenges that involve training Reinforcement Learning agents on these games. To explore the Multi-agent setting provided by PyTAG we train the popular Proximal Policy Optimisation Reinforcement Learning algorithm using self-play on a subset of games and evaluate the trained policies against some simple agents and Monte-Carlo Tree Search implemented in the Tabletop Games framework.

All of these tabletop games provide players an opportunity for turn-based strategy, which is naturally a slower, more methodical style of gameplay than open-world combat. In "Horizon Forbidden West," each totem for Machine Strike is modeled off the robotic animals you fight against in the wilds of Horizon's world. To defeat a glinthawk in "Forbidden West," you need to aim and strike the sack of chillwater in its breastbone, all while dodging to avoid the creature's incoming attacks. To win against a glinthawk totem in Machine Strike, you need to methodically outmanuever your opponent in a step-by-step process across tiles on the game board. You're allowed the time to step back and think.

From the beginning if the history of AI, there has been interest in games as a platform of research. As the field developed, human-level competence in complex games became a target researchers worked to reach. Only relatively recently has this target been finally met for traditional tabletop games such as Backgammon, Chess and Go. Current research focus has shifted to electronic games, which provide unique challenges. As is often the case with AI research, these results are liable to be exaggerated or misrepresented by either authors or third parties. The extent to which these games benchmark consist of fair competition between human and AI is also a matter of debate. In this work, we review the statements made by authors and third parties in the general media and academic circle about these game benchmark results and discuss factors that can impact the perception of fairness in the contest between humans and machines

Magic: The Gathering, which counts over 35 million players worldwide, will get an esports league with a combined $10 million in prize money, Wizards of the Coast (WOTC) and Hasbro announced. The Magic Pro League (MPL) will cover both the tabletop games and Arena digital PC version (still in public beta) and count 32 of the world's top players. Players will compete in weekly games and major "Mythic Championship" tournaments, and receive $75,000 contracts with prize money on top. The first Mythic Championship will launch during PAX East at Boston in 2019 with a $1 million purse. Entering esports will provide a few major benefits for Magic: The Gathering, its backers believe.

The world of tabletop gaming was once the preserve of nervous teenage boys holed up in dark basements and bedrooms. But as shows like Stranger Things tap into the rose-tinted nostalgia of afternoons spent playing Dungeons & Dragons, could it be that the geeks have inherited more than Middle Earth? In a craft beer bar lit with bare bulbs, a cluster of tattooed and bearded punters gather. On first glance this branch of BrewDog in Nottingham might seem like your typical hipster hangout, but one thing gives it a slightly different air: numerous hand-drawn maps, some character sheets, and voluminous bags of 20-sided dice. It's the bar's monthly tabletop gaming night - and regulars love it. "I think the escapism is the best bit," says 27-year-old gamer Hannah Yeates.