Dimensionality, granularity, and differential residual weighted entropy
- Submitting institution
-
Kingston University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-54-1383
- Type
- D - Journal article
- DOI
-
10.3390/e21090825
- Title of journal
- Entropy
- Article number
- -
- First page
- 825
- Volume
- 21
- Issue
- -
- ISSN
- 1099-4300
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Entropy, a measure of “randomness” or “unpredictability”, is an important concept in both physics and information theory, and is used as a metric in machine learning systems. Whilst differential entropy – the version required when considering a quantity which can vary continuously, rather than changing in discrete steps – can easily be defined for dimensionless quantities, there are apparent paradoxes relating to the magnitude of the units used which arise for quantities that have physical dimensions. This paper proposes some approaches towards resolving these paradoxes, using a “working granularity” to reconcile the results for discrete and continuous quantities.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -