3 ms·
"[T]he founder of modern information geometry, defined information geometry in the preface of his latest textbook [2] as follows: 'Information geometry is a met
by frob 6y ago
"[T]he founder of modern information geometry,
defined information geometry in the preface of his latest textbook [2] as follows: 'Information geometry
is a method of exploring the world of information by means of modern geometry.' In short, information
geometry geometrically investigates information sciences."
Holy circular definitions, Batman!
- eternalban 6y agoIt's not actually circular. Think of the analog example of "exploring time-space of physical reality (physics)" by the means of modern mathematics: https://en.wikipedia.org/wiki/Lorentz_transformation https://en.wikipedia.org/wiki/Lorentz_transformation So the general idea (haven't read OP yet) would be to map problems in information theory to geometric constructs, and use the machinery of modern geometry to explore that domain. This is for example using algebraic topology to applications in concurrent processing: https://www.springer.com/us/book/9783319153971 https://www.springer.com/us/book/9783319153971
- frob 6y agoThe difference between what you gave and what they did is you used different words in the definition from the description. They literally just use the same words in the name as in the definition. Their statement in your terms would be more like, "Mathematical Physics is a method of exploring the world of physics by means of modern mathematics," which is a useless statement. By no means am I saying this field is worthless or a scam, I'm just saying those definitions are completely useless and circular.