What is a dynatrope?

Dynatropes use what they have

tl;dr Dynatropes use what they have and make the best of that. They learn the limit of their gear before moving on to more gear.

Like plants are things that move towards more light, dynatropes are things that move towards more options. A dynatropism is a motion towards regions in space that provide more possibilities for more effective actions. A dynatrope is a thing performing this motion. An essential option is energy. No energy, no option whatsoever.

Definition, please

Dynatropism. From greek dynamis (power)[4] and tropos (a turning)[5]. Drive to option, Streben nach Möglichkeit oder Kraft, search for empowering states.

The dyna- part indicates power as the characteristic sensorimotor modality. The -tropos bit means ‘directed motion towards’ by the ability to turn into any direction at all.

Things that are controlled by a dynatropism are dynatropes.

That’s cool, but do we need this word?

The phenomenon is real for all practical purposes and deserves a name, of course. It has come to play an important part in my personal thinking and thus also in the projects I am in involved in, farmersmanual for one, and even more so for Jetpack. There, we are constantly searching for appropriate names and language to talk about the things that we are making, including that language itself ⥁. The naming is compatible with the biological naming system [5], and integrates seamlessly into the taxonomy of movements. It is used by the BEAM [0] robotics community to categorize different BEAMbots according to their goal seeking behavior, for example audiotropes, phototropes, radiotropes, and thermotropes.

There is a precedent use of the term ‘Dynatrope’ [7] in the universe of the TV series ‘Doctor Who’, produced by the BBC since 1963. By the account in the fandom wiki [7], they “… were powered by mental energy”.

Measuring option and empowerment

An open question in robotics research is how does autonomous development work? The type of thing we are going through, starting with the egg cell, over fetus, infant, teenager to adult and then onwards. The learning going on during all that time is immense and intricately staged. We want to understand how this works and build our machines in a similar way.

A good part of it can be accounted for with goals, how to choose goals when no one provides them to you, how to tell good goals from the too-easy or too-hard ones, when to persist and when to give up. This goal juggling is the motivation. Many interesting models have been proposed for motivation systems [citation needed]. One general approach is called intrinsic motivation, and one variant of it is the Klyubin-Polani-Nehaniv empowerment [6,citation needed,1,3], an information theoretic measure.

As such, it provides a general measuring stick for the amount of empowerment available in a region of space by ‘counting’ the number of available options. Now a map of the space can be built visiting random places and taking a measure at each place. The things that perform directed motion in this space using the map can now be called empowerment maximizing agents or intrinsically motivated agents. These names are correct but bulky and awkward.

Get the word out

Working on the scientific and technical aspects of the concept, and on the sensorimotor account of the method, are only one reason for a need of language. From our specialist point of view it is obvious that an account of the concept in everyday language terms could be quite helpful in many other areas. Seeing it as part of our mission to help create a broader understanding and the ability to apply the concept and the learnings it provides. The diffusion of several important research insights which have not made it into the general public knowledge is a larger communication task worth a post of its own [WIP].

We build bridges

We use the metaphors of building-bridges or playing-message-ferry for reflecting this activity. As explorers and scientists it is part of the job spec to do that. For some reason the task has been insufficiently attended to or even neglected by the science community at large for a long time. We think that this is part of the reason for a gaping divide in people’s and organizations’ abilities of reality-understanding.

Confounding causes with effects alone is bad enough, but teamed up with dramatic social disparity it just creates a massive vortex cloud of planet-scale annihilation options for humans and friends. Obviously very undesirable and action required.

Refs

[1] Klyubin-Polani-Nehaniv empowerment is a concept from artifical life and developmental learning research. Related concepts are intrinsic motivation, artificial motivation, autonomous learning, autonomous mental development, lifelong learning, homeokinesis, ultrastability, predictive information maximization, guided self-organization, and more. Highly recommended topic, also worth a post on its own [WIP].

[3] The everyday language meaning of empowerment corresponds quite well with whats made more precise in the mathematical model.

[0] https://en.wikipedia.org/wiki/BEAM_robotics

[4] https://en.wikipedia.org/wiki/Dynamics

[5] https://en.wikipedia.org/wiki/Tropism

[6] A. S. Klyubin, D. Polani, and C. L. Nehaniv, “All Else Being Equal Be Empowered,” in Advances in Artificial Life, vol. 3630, M. S. Capcarrère, A. A. Freitas, P. J. Bentley, C. G. Johnson, and J. Timmis, Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2005, pp. 744–753.

[7] in: Spacecraft types: Dynatrope - Dynatropes were the machine spacecraft used by the Krotons. https://tardis.fandom.com/wiki/Dynatrope

Extra detail

Tropisms basics

When things are moving around in space in any kind of way that’s motion. When things move around in a directed way thats a tropic motion. Tropic motion is generated by a tropism, a biological control system.

In perception the different channels like ears, eyes, noses, feet, and so on are called sensory modalities. Sensory modality can be used to put tropisms into different groups and then look at what is shared among all tropisms and what is unique to particular tropisms.

There is a lot of cool words for many kinds of modalities. These words are chosen to indicate the physical phenomenon that is the basis for the biochemical reactions of biological receptors, like a hair cell in the inner ear, or a photo cell on the retina. Gah, spilled it, so photo means light, chemo means molecules, hydro means water, electro means electro, everything else needs to be looked up on Wikipedia.

Now we can build a phototropism, understand that this is something that looks where the light is coming from and directs the motion of another thing doing the actual moving around. That thing is called a phototrope. Chemotropes are things that move towards or away from particular molecules. Most plants are phototropes, animals are chemotropes, people too.

TODO

  • add fm/jcl logos and fandom wiki pic as dynatrope examples
  • add reference to teleology in Context section re: BEAM goal seeking behavior
  • see new scientist, 19/26 December 2020 p.51: Vermeji argues that evolution is getting better at producing new. “Organisms have, on the whole, become more energetic over time,” he says. It is known, for instance, that the veins running through the leaves of flowering plants have become more densely packed over millions of years, meaning modern flowering plants can extract more energy from sunlight than their ancestors could 100 million years ago. Similarly, “the average metabolic rates of things like snails and clams and many more organisms have increased over time”, says Vermeji.

Changes

  • updates 2020-12-21
  • early feedback fixing names and references 2020-12-20
  • init 2020-12-19

Comments

tag cloud

robotics music AI books research psychology intelligence feed ethology computational startups sound jcl audio brain organization motivation models micro management jetpack funding dsp testing test synthesis sonfication smp scope risk principles musician motion mapping language gt fail exploration evolution epistemology digital decision datadriven computing computer-music complexity algorithms aesthetics wayfinding visualization tools theory temporal sustainability supercollider stuff sonic-art sonic-ambience society signal-processing self score robots robot-learning robot python pxp priors predictive policies philosophy perception organization-of-behavior open-world open-culture neuroscience networking network navigation movies minecraft midi message-passing measures math locomotion linux learning kpi internet init health hacker growth grounding graphical generative gaming games explanation event-representation embedding economy discrete development definitions cyberspace culture creativity computer compmus cognition business birds biology bio-inspiration android ai agents action