Exact complexity: The spectral decomposition of intrinsic computation
详细信息 查看全文
- 作者：James P. Crutchfielda ; chaos@ucdavis.edu" class="auth_mail" title="E-mail the corresponding author ; Christopher J. Ellisonb ; cellison@wisc.edu" class="auth_mail" title="E-mail the corresponding author ; Paul M. Riechersa ; pmriechers@ucdavis.edu" class="auth_mail" title="E-mail the corresponding author
- 关键词：Excess entropy ; Statistical complexity ; Projection operator ; Resolvent ; Entropy rate ; Predictable information
- 刊名：Physics Letters A
- 出版年：2016
- 出版时间：6 March 2016
- 年：2016
- 卷：380
- 期：9-10
- 页码：998-1002
- 全文大小：265 K
文摘
- •
We provide exact, closed-form expressions for a hidden stationary process' intrinsic computation.
- •
These include information measures such as the excess entropy, transient information, and synchronization information and the entropy-rate finite-length approximations.
- •
The method uses an epsilon-machine's mixed-state presentation.
- •
The spectral decomposition of the mixed-state presentation relies on the recent development of meromorphic functional calculus for nondiagonalizable operators.