Foundations of Information Science & Human Level AI

Page 1

A PRIORI MODELING OF INFORMATION Foundations of Information Science (FIS) Online Session: 16 June - 15 July 2016 (http://listas.unizar.es/pipermail/fis/) Marcus Abundis 1 This session covers the a priori modeling of information, targeting a meaningful void named by Shannon and Weaver (1949). It thus posits a universal “theory of meaning” and “theory of information” (UTI). A priori views can help as they often convey organizing principles. Also, a winning model should offer gains likely to surpass gains seen from Shannon’s (1948) prior work. For example, firm notions of meaning and universality are key to founding a human level AI and to addressing hurdles in quantum mechanics/computing and in material science (Aspuru-Guzik, 2015). An a priori view starts with what comes before information, using analytic philosophy to frame initial concepts but ends in empiric phenomenology. The session thus entails diverse levels of analysis and can stir confusion. For example, disorder at one level necessarily implies a type of order, at a different level. Different levels require a use of type theory, re Bateson’s (1979) “differences themselves must be differentiated,” to convey one system of thought. To aid this exploration a cursory model is thus offered. With the foregoing cautionary notes in mind, I invite you to join this FIS session and together we will see what unfolds (3 pages; 1,100 words). MODEL DEVELOPMENT In brief, the model synthesizes Shannon entropy, Bateson’s different differences, and Darwinian selection (S-B-D) to derive “meaningful information” in trans-disciplinary roles. The video link below details a posited model as follows: Basic issues and questions are framed (2:30 minutes). Known meaningful metadata are detailed (2:30 minutes). Next, metadata’s role is deconstructed in remaining minutes to name universal a priori facets. Lastly, the model is re-constituted “from the ground up” to present a fully synthesized S-B-D a priori view (23 minutes total). The model’s central logic follows these steps: 1) Universal meaning is named as: a thing exists (order), a thing does not exist (disorder), and a becoming-ness (diverse entropic events) that joins the two states. This triune law is labeled Generic Entropy (Δz in video). Three things are achieved here: a) Practical meaning is named for an all-inclusive cosmos; b) An open system framing is given to that Cosmic All; c) a stage is set for simple evolutionary bifurcations. This necessarily base view of universal meaning provides a foundation for further UTI analysis. 2) Universal meaning has a dualist-triune form. Each facet (existence, non-existence, and becomingness) helps define the other two, in a Bateson-like (1979) necessary unity. Three key facets (triune), in dual logical roles (existential and dynamic), thus name an initial organizing principle or topology. This topology does not mean that all information has a dualist-triune form, but that all information arises from this structural fundament. Further . . . 3) Dualist logic has many faces as order & disorder, life & death, signal & noise, good & evil, coherence & decoherence, etc. Dualism thus affords nascent complexity in respective roles 1

Organizational Behavior (GFTP), Graduate School of Business, Stanford University (March 2011).

December 31 2016, M. Abundis, +1-530.388.5576, +41-(0)62.844.2193, 55mrcs@gmail.com

1


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.