List of Posts


This blog is not really a blog as respective sites are usually called. There is no dominance of the timeline, or chronicity, despite the fact that the later posts tend to integrate the results of previous ones. Else, the posts are not as independent as it is the case in a default blog. I do not report about all the little things in one’s everyday life, except the fact, maybe, that currently I am not doing anything else except writing here and programming according to to these writings.

So, in some respect, the collection develops somehow into something that may be considered more as a book. Yet, this site does not offer a serial order of “chapters,” it (hopefully) unfolds more like a network. Else, you should not expect a textbook, or introductory notes; I assume a considerable amount of background knowledge throughout the pages, albeit it should be easy to find the respective relevant foundational sources in the web. Probably I should call it blook, allowing for any pronunciation between “blɒg” (as in blog) and “blʊk” (as in book, see about IPA here)

Anyway, here is a (more or less alphabetic) list of the posts on this site including a short description (essays about urbanism are not included so far). An index of keywords will be added later, and finally I probably will publish everything as a book.

(Note that trying to visit unpublished pages will send you to the 404 page with a duck… really!)



Future Topics in Computing thematic collections of high-level keywords
Future Topics in Philosophy ./.
Topics in Between ./.
A Model for Analogical Thinking a brief introduction into Douglas Hofstadter’s CopyCat project, that may serve as a model for creative analogical thinking.
Analogical Thinking, revisited This essay comes in two parts (2nd part); the goal of it is to establish the possibility for a crossing-over of the two paradigms of Copycat and SOM. In order to achieve that, we extensively apply the technique of elementarization, which in turn allows the synthesis of a common differential of the two quite different paradigms; based on that, some important extensions to the classical SOM paradigm are derived and introduced.
Associative Storage and Memory We first sharply distinguish database from associative storage and memory. Without dropping the separation between “storage” and “processing” the emergence of machine-based episteme will not be achievable. This equates to building a fully-fledged “memory”. We propose that memory is built on the basis of a population of probabilistic contexts in an associative structure. We discuss the WebSom (Kaski & Kohonen) as a model for associative storage. Finally, necessary components for the transition from associative storage to (true) memory are proposed.
But it Does Move. Transition from the probabilistic to the propositional; identifies the so-called “symbol grounding” as an inappropriately formulated problem, as it does for the so-called body-mind-problem;
The Category of Models In the chapter about modeling we achieved a formalization of the notion of models, yet, we also demonstrated that models can not be conceived in set-theoretic terms. Here in this chapter we show that it is more appropriate to think of models (and of data) as categories. The formal properties of mathematical categories support our perspective onto machine-based epistemology and the role of associative mechanisms in it.
Complexity By means of the technique of “elementarization” a complete and positively definite definition is provided, based on necessary and sufficient elements (compound conditions)
Conditions & Choreosteme Provides arguments about the necessity of the concept of the choreosteme (see also the “Deleuzean Move” below). That is a self-referential concept about ultimate conditions of epistemic being. We call it choreosteme and identify it as the only universal about which we can talk (so it is the only universal we can think of).
Context Provides a formal definition of “context”; the derived concept is used for routing decisions in NooLabGlue as well as in the initial preparation of textual/language input data/structures or in the generalized feature vector.
Connectivity, Levels, and Boxes Considers the level of integration for modular software systems, some tools and architectural principles
Data as far as data from some measurement can be represented by a structure apriori (which does neither apply to textual input or analogy making), the preparation of data can be automated, quite to the contrary of what is told by the mainstream; This leads us to the concept of “data experiment,” and the possibility to derive structural hypotheses about the data in a mechanic manner.
A Deleuzean Move This piece, which has been growing into sth more than a blog entry, presents a philosophical framing for mentality, and probably for consciousness. Its cornerstones are the Differential (Deleuze), the language game (Wittgenstein), transcendental entities conditioning mentality  that establish a quite unusual space for which a mathematical description is given, and a self-referential dynamics that vanishes the old problems of justification and primary causes/reasons. The four transcendental conditions are Model, Concepts, Mediality, Virtuality. The resulting space is labelled the choreostemic space (CS). The CS is then used to reconstruct notorious epistemic concepts like believing, understanding, describing or knowing in a non-reductionist manner. It is shown how the CS avoids entailment by anthropology. Finally, a critique of “Paradoxically Conditioned Reason” is being offered.We regard the CS as a main pillar for any theory of mentality, whether “human” or “machinal”.
Evolution in Associative Systems includes a formalization and hence a very general account of evolutionary processes;

The Formal and the Creative

a series of essays about the relations between formalization and creativity. Both are characterized as strongly singular terms, a concept that we use to replace the ordinary singular term as it is widely used in philosophy.  A generalization of the Wittgensteinian language games is proposed. Elementarization and Expressibility are identified to be closely related and the cornerstones of inventively describing a problematics.
Growth Describes basic (phenomenological) modes of growth, introduces abstract growth and equals it with differentiation; discusses the requirements for autonomously growing software systems
How to Glue Introduces some basic specification for a middleware (“NooLabGlue“) suitable for machine-based cognitive systems
How to Grow Associative Maps? application of insights about growth to SOMs
Ideas and Machinic Platonism We describe the process of how a SOM can be set up in order to develop own (simple) “ideas” and idealizations.
Implementing the Ability to Abstract

planned, not published.

A whole lot of self-referential problems await us here; we start with a small catalog of different kinds of abstractions, followed by a discussion of the role of symbols, signs, concepts and models in that; we also provide the architecture for an implementation of a system that is able to abstract.

Information & Causality Provides a holistic (non-reductionist) notion of information, which is related to causality;it is demonstrated that their is a complementarity between information and causality, and how both are related through observation
Interaction Extends the notion of interaction by the notion of a “behavioral coating”; not published on this site yet, but you may find a previous version here;
Language Some remarks about the role of language with regard to our epistemic and choreostemic constitution
Logic, hooked up More details about the transition from probabilistic description to propositional reference; there are three main results: (i) we have to distinguish between transcendental (T-) logic and its actualizations, the quasi (Q-) logics; (ii) any particular Q-logic is both a consequence of a certain bodilyness as well as a kind of materiality; (iii) The instantiation of a Q-logic implies a separation of temporal reference into at least two lattices. So, at the end we will transform one big question into another: How to enable for the act of naming?
Machine-based Epistemology A brief differential introduction and justification of the term
Mental States marks this term as bare nonsense
The Miracle of Comparing About the anatomy of comparing; distinction of three different types; the Deleuzean differential as a diagram; introducing the concept of ortho-regulation

in progress (96%)

The most general formal definition of a model is given, reaching far beyond mathematical model theory or the accounts known from philosophy of science

Morphology of Experience

First it is argued how the relation between experience and its technical (techné) aspects could be conceived. Then, in the second part of the essay, a detailed account of the model selection problem is given, together with its solution.

Non-Turing-Computing We argue that Non-Turing Computation (NTC) is basically equal to (sometimes adaptive) differentiation processes in bodies; we propose that NTC may be defined as computation that extends its own axiomatic basis.
NooLabGlue (Software) Description of the software that can be used to connect components of a massively modular software system, especially if those modules are artificial associative structures such as Self-Organizing Maps or Artificial Neural Networks
Probabilistic Networks

not published.

Networks without borders and without idealistic relations; introduces the notion of “randolation” that denotes the probabilistic sibling of the relation; compares reaction-diffusion-systems and self-organizing maps

Representation We conclude that the most appropriate input to the association engine are probabilistic distributions upon arbitrarily chosen “basic” properties

The Self-Organizing Map –
an Intro

just that.

The Self-Organizing Map:
SomeDesign Issues

We identify key elements of the SOM and discuss possible design decisions for creating a SOM, the weaknesses of commonly applied settings as well as possible improvements.
Similarity It starts with a brief survey about the usage of the concept in various domains; with respect to similarity it is concluded that statistics is a black box methodology and hence seriously unsuitable for predictive modeling; Euclidean distance is identified as inappropriate operationalization of similarity; identity is sharply distinguished from similarity, both concepts are seen as complementary sides of the inevitable transition from probabilistic description to propositional reference; it is shown why similarity is not an ontological category, but nothing else than a (structural) hypothesis; finally, we arrive at the remarkable generalization according to which similarity can be conceived as a mapping of relations onto a scalar.
The SOM and the Symbolic Solving the puzzles around the genesis of symbols in formal or material processes;
The (extended) SOM Mechanism

not published.

A detailed exhibition of the rules for creating an improved and accelerated self-organizing map

Technical Aspects of Modeling Issues of implementation: how to arrange and organize the various elements of modeling, how to prepare data, how to (re)present them, how to reason about the difference of models, things not to do in modeling, some nonsensical, though abundant, habits in modeling
Theory (of Theory) It is argued that “theory” (1) is a particular language game, and (2) should be conceived as a (ortho-regulative) milieu of modeling. As such theories are not concerned with any kind of empirical content such as predictions. Moreover, there is no formal relation between a theory and the models built from/with it.
Vagueness Following and largely agreeing with an analysis first proposed by Bertrand Russell, we dismiss the stance according to which the notion of vagueness is problematic or even paradoxical; we also refute mainstream linguistic analysis of adjectives by means of logics or other formalisms. Else, many linguists claim a weird relationship between language and model, leaving the status of model ambivalent. In order to acknowledge and embrace the vagueness of language as a positive design principle, rather than a kind of aberration, we propose a novel structural and necessary element for words, the processual indicative, which can be conceived as a structural and “technical” extension of inferentialism.
Waves, Words and Images Here we investigate the link between perceptions and words, the twofold question around representation and the role of words in higher mental capabilities.
Collected References



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: