Computational Mechanics: Why Model?
Structural Complexity of Information Source
C_ = H[Pr(S)], S = {Casual states}
Uses:
Environíl complexity: Amount/kind of relevant structure
Agentís inferential capacity: Sophistication of models?
Theorem: E _ C_
Conclusion: Build models vs. storing only E bits of history.
Raw sequence blocks do not allow optimal prediction,
only E bits of mutual information in blocks.
Optimal prediction requires larger model: 2C_, not 2E.
Explicit: 1D Range-R Ising spin system: C_ =E+Rh_.
Previous slide
Next slide
Back to first slide
View graphic version