• Corpus ID: 211069096

Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales

@article{Earle2020UsingFN,
  title={Using Fractal Neural Networks to Play SimCity 1 and Conway's Game of Life at Variable Scales},
  author={Sam Earle},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.03896}
}
  • Sam Earle
  • Published 29 January 2020
  • Computer Science
  • ArXiv
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… 

Figures from this paper

Illuminating diverse neural cellular automata for level generation

A quality diversity (QD) approach to generating a collection of NCA level generators by framing the problem as a QD problem, which can train diverse level generators, whose output levels vary based on aesthetic or functional criteria.

Say "Sul Sul!" to SimSim, A Sims-Inspired Platform for Sandbox Game AI

Environment design in the life simulation game The Sims is proposed as a novel platform and challenge for testing divergent search algorithms and the ability of novelty-based evolutionary algorithms to effectively generate viable environment designs is investigated.

IORand: A Procedural Videogame Level Generator Based on a Hybrid PCG Algorithm

The intelligent orchestrator of random generators (IORand), a hybrid procedural content generation (PCG) algorithm, driven by game experience, based on reinforcement learning and semi-random content generation methods is presented.

Automatic Game Level Generation

This work explores the novel (as to the authors' knowledge) formulation of procedural level generation (PLG) as a task of learning a mapping from the difficulty to the level embedding space, using an N-dimensional seeded generator.

Exploring open-ended gameplay features with Micro RollerCoaster Tycoon

Results indicate that building from scratch with no costs results in the widest diversity of high-performing designs.

Learning Controllable Content Generators

This work proposes to train generators capable of producing controllably diverse output, by making them “goal-aware,” by adding conditional inputs representing how close a generator is to some heuristic, and also modifying the reward mechanism to incorporate that value.

PCGRL: Procedural Content Generation via Reinforcement Learning

Three different ways of transforming two-dimensional level design problems into Markov decision processes are investigated, and these are applied to three game environments.

Deep learning for procedural content generation

This article surveys the various deep learning methods that have been applied to generate game content directly or indirectly, discusses deeplearning methods that could be used for content generation purposes but are rarely used today, and envisages some limitations and potential future directions of deep learning for procedural content generation.

Multi-Objective level generator generation with Marahel

This paper uses NSGA-II, a multi-objective optimization algorithm, to search for generators for three different problems (Binary, Zelda, and Sokoban), and restricts the representation to a subset of Marahel language to push the evolution to find more efficient generators.

Video Games as a Testbed for Open-Ended Phenomena

Results indicate that experiments with SimCity are more likely to identify statistically significant differences in complexity as a result of applied destructive events, highlighting the utility of this game domain for studying artificial life phenomena.

References

SHOWING 1-10 OF 11 REFERENCES

Mastering the game of Go with deep neural networks and tree search

Using this search algorithm, the program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0.5, the first time that a computer program has defeated a human professional player in the full-sized game of Go.

StarCraft II: A New Challenge for Reinforcement Learning

This paper introduces SC2LE (StarCraft II Learning Environment), a reinforcement learning environment based on the StarCraft II game that offers a new and challenging environment for exploring deep reinforcement learning algorithms and architectures and gives initial baseline results for neural networks trained from this data to predict game outcomes and player actions.

New constructions in cellular automata

This book presents self-Organized Constructions in Sparse Random Arrays of Conway's Game of Life as well as a Two-Dimensional Cellular Automaton Crystal with Irrational Density, a model for simulating Digital Logic with the Reversible Aggregation Model of Crystal Growth.

The simulation and prediction of spatio-temporal urban growth trends using cellular automata models: A review

Cellular Automata and Fractal Urban Form: A Cellular Modelling Approach to the Evolution of Urban Land-Use Patterns

A cellular automaton is developed to model the spatial structure of urban land use over time and to link the results directly to general theories of structural evolution.

FractalNet: Ultra-Deep Neural Networks without Residuals

In experiments, fractal networks match the excellent performance of standard residual networks on both CIFAR and ImageNet classification tasks, thereby demonstrating that residual representations may not be fundamental to the success of extremely deep convolutional neural networks.

Asynchronous Methods for Deep Reinforcement Learning

A conceptually simple and lightweight framework for deep reinforcement learning that uses asynchronous gradient descent for optimization of deep neural network controllers and shows that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task of navigating random 3D mazes using a visual input.

The Semiotics of SimCity

This paper explores SimCity and its significance as a simulator not only of reality but consciousness, discovering how software is organized for a certain set of goals and actions.

Fractal Cities: A Geometry of Form and Function

Planners and architects will find that many aspects of fractal geometry covered in this book are relevant to their own interests.

New games of life : Cellular automata and subsurface discourses in simcity

  • 2011