ᛕᎥᕼᗷᗴᖇᑎᗴ丅Ꭵᑕᔕ @Kihbernetics@qoto.org

Found it! 😎. It is called the
***Maximum Entropy Production Principle (#MEPP)***:

>The so-called maximum entropy production principle (MEPP) is known much less (even among specialists in physics of nonequilibrium processes). This antipode, as its name seemingly means, of Prigogine's principle has been overshadowed by its more famous twin. MEPP was independently proposed and used by several scientists throughout the 20th century when they dealt with general theoretical issues of thermodynamics and statistical physics or solved specific problems. By this principle, a nonequilibrium system develops so as to maximize its entropy production under present constraints.

https://www.sciencedirect.com/science/article/abs/pii/S0370157305004813

A system doesn't ***feed*** on #order (or #negentropy) from the environment it has to ***create*** it.

You can't get your desk organized by just acquiring some order from the environment. You have to do some #work and use some of your #free_energy. Schrödinger admits as much:

#Life as the ultimate form of #self_organization is usually thought of as ***resisting*** the *Second Law* of thermodynamics that says all #order eventually dissipates over time into a state of thermal equilibrium and enduring uniformity (disorder).

An alternative explanation says that organization (order) spontaneously emerges in dissipative structures ***because*** of the Second Law of Thermodynamics because ordered structures are much better at dissipating energy (thus more rapidly increasing the #entropy) than disordered ones.

The explanation (for which I now can't find the proper reference😟) exemplified this with whirlpools and how they spontaneously emerge because the water molecules in them don't bump into each other as much so the flow through the drain (transition to an equilibrium state) is faster when they are streamlined (organized) into a vortex.

@psybertron @tg9541

Cybernetics has a peculiar relationship with the topic of self-organization. I don't recall Wiener having ever taken this matter seriously, von Foerster thought a system is "feeding" on the order from the environment, and this is Ashby talking about "self" organization requiring the presence of another machine.
That's why I had to "invent" Kihbernetics and return to the basics.😀

@tg9541 @psybertron

That's precisely the point I'm trying to make. Any mindset that is based on the distinction between the #controlled and the #control system, must be very attractive to would-be dictators.

@tg9541 @psybertron

Yes. I think the emergence of modern computing in the same period didn't help cybernetics at all because everything was (and still is) considered as "computation", although the ever-increasing power of computation, as Ashby noted, allowed the analysis and simulation of "medium complexity" systems along with traditional deterministic and stochastic methods for dealing with the extreme cases.

@psybertron @tg9541

Agree. I think Ashby's "noble" objective should be the real purpose:

>Cybernetics offers one set of concepts that, by having exact
correspondences with each branch of science can thereby bring them into exact relation with one other.

But then many, including Wiener, had a narrow misguided view when worrying about the "wrong use" of cybernetic automation in "enslaving" humanity.

@psybertron @tg9541

I would be very interested in any reference to examples of how to use of cybernetic thinking "to understand how to guard against totalitarian/amoral/ideological (control) purposes".

#Trust is #Experience over #Time.

@tg9541

#AI, like all other machines we invented, produced, and use in our daily lives is an #amplifier of our ability to #reason, in the same way that cybernetic #automation was (is) an amplifier of our ability to better (with more power) #control our environment, and radios, computers, the #internet, and social media in particular, are all amplifiers of our #communication capabilities.

Amplifiers do not care what they "amplify", so if you have "garbage in" you can be assured that you'll get much more "garbage out", and because amplifiers are very sensitive to ***positive*** (reinforcing) #feedback, you need some reliable ***negative*** (regulation) feedback to keep them in check.

#AI and #Trust

>I trusted a lot today. I trusted my phone to wake me on time. I trusted Uber to arrange a taxi for me, and the driver to get me to the airport safely. I trusted thousands of other drivers on the road not to ram my car on the way. At the airport, I trusted ticket agents and maintenance engineers and everyone else who keeps airlines operating. And the pilot of the plane I flew in. And thousands of other people at the airport and on the plane, any of which could have attacked me. And all the people that prepared and served my breakfast, and the entire food supply chain—any of them could have poisoned me. When I landed here, I trusted thousands more people: at the airport, on the road, in this building, in this room. And that was all before 10:30 this morning.

https://www.schneier.com/blog/archives/2023/12/ai-and-trust.html

@tg9541 @psybertron

Semantics, information, meaning, systems, models, representations, purpose ... those are all categories we "invented" to explain what we ***think*** to other thinkers.

The primary concern in Cybernetics and automation always was (and still is) #control of things within its environment, with representation, modeling, or #explanation used only to the extent that serves this primary function.

Now, an agent doesn't need to "understand" the "semantics" of what they are doing. Agents like living cells, for example, just have to respond correctly to the syntax of the genetic message, like any other *mechanism*, to jointly produce a more complex organism agent able to "understand" and communicate what's going on and not necessarily for control purposes.

@tg9541

>Kennedy was assassinated in 1963, but in his first State of the Union the following January, Johnson urged passage of the pending Kennedy automation commission bill. Congress obliged, and on Aug. 19, 1964, LBJ signed the legislation that formally created the National Commission on Technology, Automation, and Economic Progress, noting, ***“If we have the brainpower to invent these machines, we have the brainpower to make certain that they are a boon, not a bane, to humanity.”***

😀

https://www.washingtonpost.com/history/2023/10/29/automation-technology-fear-artificial-intelligence/

@janhoglund

Nothing is perfect. The #system (and consequently the #model built upon it) is always a selected subset of all the #variables available for the #control of the #machine.

#WR_Ashby in "Design for a Brain" (chapter 2/5) explains in more detail the meaning of the word "system":

>Because any real 'machine' has an infinity of variables, from which different observers (with different aims) may reasonably make an infinity of different selections, there must first be given an observer (or experimenter); a system is then defined as any set of variables that he selects from those available on the real ‘machine’. It is thus a list, nominated by the observer, and is quite different in nature from the real ‘machine’. Throughout the book, ‘the system’ will always refer to this abstraction, not to the real material ‘machine’.

As you can never control **all** the variables of any given "machine", there is always the possibility that the variables that you don't control will generate some unforeseen consequences.

1️⃣ Kihbernetic #System with
2️⃣ fundamental #Processes: a recursive #Autopoietic self-production for growth and learning, and a linear #Allopoietic production of "other things", such as behavior and waste, distributed in
3️⃣ Control #Levels, of #Regulation, immersed in, and dealing with things in the system's environment, #Control for managing the workload of different regulators, and #Guidance to provide long-term goals and preserve the identity of the system, all using
4️⃣ #Variables: sensory #Input of data and other resources, motor #Output of behavior, #Information as the difference that will make a difference in the subsequent (updated) #Knowledge state, all interconnecting
5️⃣ #Functions: the #Control-ed #Reaction to external stimuli, the #Perception of sensory states, the #Prediction of the expected outcome of past behavior, and the repeated #Integration of new information into an updated knowledge state.

>"#N_Wiener’s preoccupation with mathematics, echoed in #WR_Ashby’s commitment to mechanical models and #Hv_Foerster’s conception of formal descriptions, largely excluded social phenomena in which cybernetics was practiced."

#K_Krippendorff's last paper

https://constructivist.info/19/1/082.krippendorff

A 3D #space of a natural #system can be visualized as created from the interaction of 2D #planes made of #matter and of #energy that modulate (change the form of or constrain) each other, leaving a 1D #information #sequence as a record of how their interaction unfolds in #time.

All #learning must be open-ended. The learning agent (the #observer) must have the #autonomy to set its own learning goals as well as plan and execute a #sequence of #exploration activities to achieve these goals.

One can never learn *all existing data* but rather refine their understanding of the data that is available to them. As true for human intelligence, you can either have "deep and narrow" specialized #AI agents or "average and broad" #AGI. You can't have both in the same entity. Time and #memory "limitation" are the main inspirations for #diversity and #cooperation between learning agents.

People should have figured it out by now that the #distribution of processing power, not the #centralization in gargantuan data and control centers is the right thing to do.

Stop working on LLMs (Large Language Models) and start working on PCAs (Personal Customizable Assistants).

From: https://arxiv.org/pdf/2311.00344.pdf