Category Archives: inference engine

OKSIMO APPLICATIONS – Simple Examples – Citizens of a County

eJournal: uffmm.org ISSN 2567-6458

27.March 2022 – 27.March 2022
Email: info@uffmm.org
Author: Gerd Doeben-Henisch
Email: gerd@doeben-henisch.de

BLOG-CONTEXT

This post is part of the Oksimo Application theme which is part of the uffmm blog.

PREFACE

This post shows a simple simulation example with the beta-version of the new Version 2 of the oksimo programming environment. This example shall illustrate the concept of an ‘Everyday Empirical Theory‘ as described in this blog 11 days before. It is intentionally as ‘simple as possible’. Probably some more examples will be shown.

FROM THEORY TO AN APPLICATION

To apply a theory concept in an everyday world there are many formats possible. In this text it will be shown how such an application would look like if one is applying the oksimo programming environment. Until now there exists only a German Blog (oksimo.org) describing the oksimo paradigm a little bit. But the examples there are written with oksimo version 1, which didn’t allow to use math. In version 2 this is possible, accompanied by some visual graph features.

Everyday Experts – Basic Ideas

This figure shows a simple outline of the basic assumptions of the oksimo programming environment constituting the oksimo paradigm: (i) Every human person is assumed to be a ‘natural expert’ being member of a bigger population which shares the same ‘everyday language’ including basic math. (ii) An actor is embedded in some empirical environment including the own body and other human actors. (iii) Human actors are capable of elaborating as inner states different kinds of ‘mental (cognitive) models’ based on their experience of the environment and their own body. (iv) Human actors are further capable to use symbolic languages to ‘represent’ properties of these mental models encoded in symbolic expressions. Such symbolic encoding presupposes an ‘inner meaning function’ which has to be learned. (v) In the oksimo programming environment one needs for the description of a ‘given state’ two kinds of symbolic expressions: (v.1) Language expressions to describe general properties and relations which are assumed to be ‘given’ (= ‘valid by experience’); (v.2) Language expressions to name concrete quantitative properties (simple math expressions).

This figure shows the idea how to change a given state (situation) by so-called ‘change rules’. A change rule encodes experience from the everyday world under which conditions some properties of a given situation S can be ‘changed’ in a way, that a ‘new situation’ S* comes into being. Generally a given state can change if either language expression is ‘deleted’ from the description or ‘contributed’. Another possibility is realized if one of the given quantitative expressions changes its value. In the above simple situation the only change happens by changing the number of citizens by some growth effect. But, as other examples will demonstrate, everything is possible what is possible in the empirical world.

SOME MORE FEATURES

The basic schema of the oksimo paradigm assumes the following components:

  1. The description of a ‘given situation’ as a ‘start state’.
  2. The description of a ‘vision’ functioning as a ‘goal’ which allows a basic ‘Benchmarking’.
  3. A list of ‘change rules’ which describe the assumed possible changes
  4. An ‘inference engine’ called ‘simulator’: Depending from the number of wanted ‘simulation cycles’ (‘inferences’) the simulator applies the change rules onto a given state S and thereby it is producing a ‘follow up state’ S*, which becomes the new given state. The series of generated states represents the ‘history’ of a simulation. Every follow up state is an ‘inference’ and by definition also a ‘forecast’.

All these features (1) – (4) together constitute a full empirical theory in the sense of the mentioned theory post before.

Let us look to a real simulation.

A REAL SIMULATION

The following example has been run with Oksimo v2.0 (Pre-Release) (353e5). Hopefully we can finish the pre-release to a full release the next few weeks.

A VISION

Name: v2026

Expressions:

The Main-Kinzig County exists.

Math expressions:

YEAR>2025 and YEAR<2027

This simple goal assumes the existence of the Main-Kinzig County for the year 2026.

GIVEN START STATE

Name: StartSimple1

Expressions:

The Main-Kinzig County exists.

The number of citizens is known.

Comparing the number of different years one has computed a growth rate.

Math expressions:

YEAR=2018Number

CITIZENS=418950Amount

GROWTH=0.0023Percentage

The start state makes some simple statements which are assumed to be ‘valid’ in a ‘real given situation’ by the participating natural experts.

CHANGE RULES

In this example there is only one change rules (In principle there can be as many change rules as wanted).

Rule name: Growth1

Probability: 1.0

Conditions:

The Main-Kinzig County exists.

Math conditions:

CITIZENS < 450000

Effects plus:

Effects minus:

Effects math:

CITIZENS=CITIZENS+(CITIZENS*GROWTH)

YEAR=YEAR+1

This change rules is rather simple. It looks only to the fact whether the Main-Kinzig County exists and wether the number of citizens is still below 450000. If this is the case, then the year will be incremented and the number of citizens will be incremented according to an extremely simple formula.

For every named quantity in this simulation (YEAR, GROWTH, CITIZENS) the values are collected for every simulation cycle and therefore can be used for evaluations. In this simple case only the quantities of YEAR and CITIZENS have changes:

Simple linear graph for the quantity named YEAR
Simple linear graph for the quantity named CITIZENS

Here the quick log of simulation cycle round 7 – 9:

Round 7

State rules:
Vision rules:
Current states: The number of citizens is known.,Comparing the number of different years one has computed a growth rate.,The Main-Kinzig County exists.
Current visions: The Main-Kinzig County exists.
Current values:
YEAR: 2025Number
CITIZENS: 425741.8149741673Amount
GROWTH: 0.0023Percentage

50.00 percent of your vision was achieved by reaching the following states:
The Main-Kinzig County exists.,
And the following math visions:
None

Round 8

State rules:
Vision rules:
Current states: The number of citizens is known.,Comparing the number of different years one has computed a growth rate.,The Main-Kinzig County exists.
Current visions: The Main-Kinzig County exists.
Current values:
YEAR: 2026Number
CITIZENS: 426721.0211486079Amount
GROWTH: 0.0023Percentage

100.00 percent of your vision was achieved by reaching the following states:
The Main-Kinzig County exists.,
And the following math visions:
YEAR>2025 and YEAR<2027,

Round 9

State rules:
Vision rules:
Current states: The number of citizens is known.,Comparing the number of different years one has computed a growth rate.,The Main-Kinzig County exists.
Current visions: The Main-Kinzig County exists.
Current values:
YEAR: 2027Number
CITIZENS: 427702.4794972497Amount
GROWTH: 0.0023Percentage

50.00 percent of your vision was achieved by reaching the following states:
The Main-Kinzig County exists.,
And the following math visions:
None

In round 8 one can see, that the simulation announces:

100.00 percent of your vision was achieved by reaching the following states: The Main-Kinzig County exists., And the following math visions: YEAR>2025 and YEAR<2027

From this the natural expert can conclude that his requirements given in the vision are ‘fulfilled’/’satisfied’.

WHAT COMES NEXT?

In a loosely order more examples will follow. Here you find the next one.