| | |
| | |
Stat |
Members: 3667 Articles: 2'599'751 Articles rated: 2609
18 February 2025 |
|
| | | |
|
Article overview
| |
|
Backward transfer entropy: Informational measure of non-hidden Markov property and its interpretation in thermodynamics, gambling and causality | Sosuke Ito
; | Date: |
1 May 2016 | Abstract: | The transfer entropy is a well-established measure of the information flow,
which has been shown useful not only in the time series analysis but also in
nonequilibrium physics. Here, we introduce the transfer entropy of the backward
path, called the backward transfer entropy, which quantifies non-hidden Markov
property. Thermodynamically, the backward transfer entropy can be interpreted
as a thermodynamic loss, where the transfer entropy implies a possible
thermodynamic benefit. In a rather different setting, the backward transfer
entropy also quantifies a loss of the gambler’s wealth of the gambling with
side information, where the transfer entropy implies a possible benefit of the
gambler’s wealth. We show that the backward transfer entropy represents the
anti-causality in the context of the causality test. | Source: | arXiv, 1605.0301 | Services: | Forum | Review | PDF | Favorites |
|
|
No review found.
Did you like this article?
Note: answers to reviews or questions about the article must be posted in the forum section.
Authors are not allowed to review their own article. They can use the forum section.
|
| |
|
|
|