Andy Nealen

Generating heuristics for novice players

We consider the problem of generating compact sub-optimal game-playing heuristics that can be understood and easily executed by novices. In particular, we seek to find heuristics that can lead to good play while at the same time be expressed as fast and frugal trees or short decision lists. This has applications in automatically generating tutorials and instructions for playing games, but also in analyzing game design and measuring game depth. We use the classic game Blackjack as a testbed, and compare condition induction with the RIPPER algorithm, exhaustive-greedy search in statement space, genetic programming and axis-aligned search. We find that all of these methods can find compact well-playing heuristics under the given constraints, with axis-aligned search performing particularly well.

Hyper-heuristic general video game playing

In general video game playing, the challenge is to create agents that play unseen games proficiently. Stochastic tree search algorithms, like Monte Carlo Tree Search, perform relatively well on this task. However, performance is non-transitive: different agents perform best in different games, which means that there is not a single agent that is the best in all the games. Rather, some types of games are dominated by a few agents whereas other different agents dominate other types of games. Thus, it should be possible to construct a hyper-agent that selects from a portfolio, in which constituent sub-agents will play a new game best. Since there is no knowledge about the games, the agent needs to use available features to predict the most suitable algorithm. This work constructs such a hyper-agent using the General Video Game Playing Framework (GVGAI). The proposed method achieves promising results that show the applicability of hyper-heuristics in general video game playing and related tasks.

SeekWhence a retrospective analysis tool for general game design

This paper describes the design of SeekWhence, a retrospective analysis tool for gameplay session. SeekWhence is a new addition to the Cicero AI-assisted game design tool, which is built on top of the Video Game Description Language (VGDL) and the General Video Game Framework (GVG-AI). With SeekWhence, designers can prototype their games and record gameplay sessions simulated by agents or human players. They can go back and forth on every frame of the recorded session, analyzing it step by step and import it into their current project to edit it. This paper explains the technical details of SeekWhence and gives examples of its usage.

Generating beginner heuristics for simple texas hold'em

Beginner heuristics for a game are simple rules that allow for effective playing. A chain of beginner heuristics of length N is the list of N rules that play the game best. Finding beginner heuristics is useful both for teaching a novice to play the game well and for understanding the dynamics of the game. We present and compare methods for finding beginner heuristics in a simple version of Poker: Pre-Flop Heads-Up Limit Texas Holdem. We find that genetic programming outperforms greedy-exhaustive search and axis-aligned search in terms of finding well-playing heuristic chains of given length. We also find that there is a limited amount of non-transitivity when playing beginner heuristics of different lengths against each other, suggesting that while simpler heuristics are somewhat general, the more complex seem to overfit their training set.

AI-based playtesting of contemporary board games

Ticket to Ride is a popular contemporary board game for two to four players, featuring a number of expansions with additional maps and tweaks to the core game mechanics. In this paper, four different game-playing agents that embody different playing styles are defined and used to analyze Ticket to Ride. Different playing styles are shown to be effective depending on the map and rule variation, and also depending on how many players play the game. The performance profiles of the different agents can be used to characterize maps and identify the most similar maps in the space of playstyles. Further analysis of the automatically played games reveal which cities on the map are most desirable, and that the relative attractiveness of cities is remarkably consistent across numbers of players. Finally, the automated analysis also reveals two classes of failures states, where the agents find states which are not covered by the game rules; this is akin to finding bugs in the rules. We see the analysis performed here as a possible template for AI-based playtesting of contemporary board games.

Talakat: bullet hell generation through constrained map-elites

We describe a search-based approach to generating new levels for bullet hell games, which are action games characterized by and requiring avoidance of a very large amount of projectiles. Levels are represented using a domain-specific description language, and search in the space defined by this language is performed by a novel variant of the Map-Elites algorithm which incorporates a feasible-infeasible approach to constraint satisfaction. Simulation-based evaluation is used to gauge the fitness of levels, using an agent based on best-first search. The performance of the agent can be tuned according to the two dimensions of strategy and dexterity, making it possible to search for level configurations that require a specific combination of both. As far as we know, this paper describes the first generator for this game genre, and includes several algorithmic innovations.

AtDELFI: automatically designing legible, full instructions for games

This paper introduces a fully automatic method for generating video game tutorials. The AtDELFI system (Automatically DEsigning Legible, Full Instructions for games) was created to investigate procedural generation of instructions that teach players how to play video games. We present a representation of game rules and mechanics using a graph system as well as a tutorial generation method that uses said graph representation. We demonstrate the concept by testing it on games within the General Video Game Artificial Intelligence (GVG-AI) framework; the paper discusses tutorials generated for eight different games. Our findings suggest that a graph representation scheme works well for simple arcade style games such as Space Invaders and Pacman, but it appears that tutorials for more complex games might require higher-level understanding of the game than just single mechanics.

Simulating strategy and dexterity for puzzle games

We examine the impact of strategy and dexterity on video games in which a player must use strategy to decide between multiple moves and must use dexterity to correctly execute those moves. We run simulation experiments on variants of two popular, interactive puzzle games: Tetris, which exhibits dexterity in the form of speed-accuracy time pressure, and Puzzle Bobble, which requires precise aiming. By modeling dexterity and strategy as separate components, we quantify the effect of each type of difficulty using normalized mean score and artificial intelligence agents that make human-like errors. We show how these techniques can model and visualize dexterity and strategy requirements as well as the effect of scoring systems on expressive range.

Evolving maps and decks for ticket to ride

We present a search-based approach to generating boards and decks of cards for the game Ticket to Ride. Our evolutionary algorithm searches for boards that allow for a well-shaped game arc, and for decks that promote an equal distribution of desirability for cities. We show examples of two boards generated by our algorithm and compare our results to those of the actual components of the game. Our approach creates game content that is specifically designed towards metrics that can affect gameplay in an impactful way.

Drawing without replacement as a game mechanic

We introduce several deck of cards and dice models that can be used to represent stochastic outcomes in tabletop games. We analyze these using a toy game introduced as a Micro Combat game. By simulating the outcome of the game with these different models we can analyze them in terms of their salience, disparity, fairness and obfuscation. We expect this analysis to help designers choose the method that best suits their intended experience.