Corpus ID: 214623314

Neural Game Engine: Accurate learning ofgeneralizable forward models from pixels

@article{Bamford2020NeuralGE,
  title={Neural Game Engine: Accurate learning ofgeneralizable forward models from pixels},
  author={Chris Bamford and Simon Lucas},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.10520}
}
  • Chris Bamford, Simon Lucas
  • Published in ArXiv 2020
  • Computer Science
  • Access to a fast and easily copied forward model of a game is essential for model-based reinforcement learning and for algorithms such as Monte Carlo tree search, and is also beneficial as a source of unlimited experience data for model-free algorithms. Learning forward models is an interesting and important challenge in order to address problems where a model is not available. Building upon previous work on the Neural GPU, this paper introduces the Neural Game Engine, as a way to learn models… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 28 REFERENCES

    Recurrent Environment Simulators

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Neural GPUs Learn Algorithms

    VIEW 14 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning and Querying Fast Generative Models for Reinforcement Learning

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Imagination-Augmented Agents for Deep Reinforcement Learning

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    A Local Approach to Forward Model Learning: Results on the Game of Life Game

    VIEW 1 EXCERPT

    An investigation of model-free planning

    VIEW 2 EXCERPTS

    Learning Latent Dynamics for Planning from Pixels

    VIEW 1 EXCERPT

    Learning Local Forward Models on Unforgiving Games

    VIEW 3 EXCERPTS