用分形神经网络玩变尺度下的SimCity 1和Conway生命游戏
Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales
摘要:
我们介绍了gym-city,一个强化学习环境,它使用SimCity 1的游戏引擎来模拟城市环境,其中代理可能会在各种大小的游戏板上寻求优化一个或任意数量的城市范围指标的组合。我们专注于人口,并分析我们的代理泛化到比训练期间看到的更大的地图大小的能力。环境是交互式的,允许人类玩家在训练和推理过程中与代理一起构建,可能会影响他们的学习过程,或者手动探测和评估他们的表现。为了测试我们的代理捕捉游戏板元素之间的距离不可知关系的能力,我们在环境中设计了一个迷你游戏,通过设计,在足够大的规模下,严格的本地策略是不可解的。鉴于游戏引擎广泛使用元胞自动机,我们还训练我们的代理“玩”康威的生命游戏-再次优化人口-并在多个尺度上检查他们的行为。为了使我们的模型与可变规模的游戏兼容,我们使用具有递归权重和结构的神经网络-分形在不同的深度被截断,这取决于游戏板的大小。
英文摘要:
We introduce gym-city, a Reinforcement Learning environment that uses SimCity 1's game engine to simulate an urban environment, wherein agents might seek to optimize one or a combination of any number of city-wide metrics, on gameboards of various sizes. We focus on population, and analyze our agents' ability to generalize to larger map-sizes than those seen during training. The environment is interactive, allowing a human player to build alongside agents during training and inference, potentially influencing the course of their learning, or manually probing and evaluating their performance. To test our agents' ability to capture distance-agnostic relationships between elements of the gameboard, we design a minigame within the environment which is, by design, unsolvable at large enough scales given strictly local strategies. Given the game engine's extensive use of Cellular Automata, we also train our agents to "play" Conway's Game of Life -- again optimizing for population -- and examine their behaviour at multiple scales. To make our models compatible with variable-scale gameplay, we use Neural Networks with recursive weights and structure -- fractals to be truncated at different depths, dependent upon the size of the gameboard.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。