Analysis of Agent Expertise in Ms. Pac-Man Using Value-of-Information-Based Policies

@article{Sledge2019AnalysisOA,
  title={Analysis of Agent Expertise in Ms. Pac-Man Using Value-of-Information-Based Policies},
  author={Isaac J. Sledge and J. Pr{\'i}ncipe},
  journal={IEEE Transactions on Games},
  year={2019},
  volume={11},
  pages={142-158}
}
  • Isaac J. Sledge, J. Príncipe
  • Published 2019
  • Computer Science, Mathematics
  • IEEE Transactions on Games
  • Conventional reinforcement-learning methods for Markov decision processes rely on weakly guided, stochastic searches to drive the learning process. It can therefore be difficult to predict what agent behaviors might emerge. In this paper, we consider an information-theoretic cost function for performing constrained stochastic searches that promote the formation of risk-averse to risk-favoring behaviors. This cost function is the value of information, which provides the optimal tradeoff between… CONTINUE READING

    Figures and Topics from this paper.

    Texture segmentation by multiscale aggregation of filter responses and shape elements
    181
    DETERMINATION OF RUBIDIUM IN SEAWATER ’
    Temporal dynamics of uncultured viruses: a new dimension in viral diversity
    17
    Value of Information Analysis via Active Learning and Knowledge Sharing in Error-Controlled Adaptive Kriging

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 63 REFERENCES
    Algebraic multigrid theory: The symmetric case
    358
    The Sedona Conference Best Practices Commentary on the Use of Search and Information Retrieval Methods in E-Discovery cautions
    16
    Efficient engraftment of human primary breast cancer transplants in nonconditioned NOD/Scid mice
    42
    From logical omniscience to partial logical competence
    1