Information Bottlenecks, Causal States, and Statistical Relevance Bases: How to Represent Relevant Information in memoryless transduction

@article{Shalizi2002InformationBC,
  title={Information Bottlenecks, Causal States, and Statistical Relevance Bases: How to Represent Relevant Information in memoryless transduction},
  author={C. Shalizi and J. Crutchfield},
  journal={Adv. Complex Syst.},
  year={2002},
  volume={5},
  pages={91-96}
}
  • C. Shalizi, J. Crutchfield
  • Published 2002
  • Mathematics, Physics, Computer Science
  • Adv. Complex Syst.
  • Discovering relevant, but possibly hidden, variables is a key step in constructing useful and predictive theories about the natural world. This brief note explains the connections between three approaches to this problem: the recently introduced information-bottleneck method, the computational mechanics approach to inferring optimal models, and Salmon's statistical relevance basis. 

    Topics from this paper.

    Information Bottleneck Approach to Predictive Inference
    • 42
    • Open Access
    Information Flows in Causal Networks
    • 194
    • Open Access
    Information theory and learning: a physical approach
    • 13
    • Open Access
    Informational and Causal Architecture of Discrete-Time Renewal Processes
    • 18
    • Open Access
    Predictability, Complexity, and Learning
    • 17
    • Open Access
    An Algorithm for Pattern Discovery in Time Series
    • 87
    • Open Access
    Information Theory of Decisions and Actions
    • 197
    • Open Access

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 20 REFERENCES
    Inferring statistical complexity.
    • 813
    Causation, prediction, and search
    • 3,340
    • Open Access
    Statistical explanation & statistical relevance
    • 317
    The calculi of emergence: computation, dynamics and induction
    • 679
    • Open Access
    Scientific Explanation and the Causal Structure of the World.
    • 847
    • Open Access
    Thermodynamic depth of causal states: Objective complexity via minimal representations
    • 119
    • Open Access
    The information bottleneck method
    • 1,883
    • Highly Influential
    • Open Access
    Statistical Explanation and Statistical Relevance
    • 98
    The Mathematical Theory of Communication
    • 10,821
    • Open Access