Monday, 26 November 2018

Why has my blogging slowed down? (Some thoughts on holograms, music and machine learning)

I realise that I haven't blogged very much recently. Partly it's because I've been very busy and a bit exhausted. But also I think it's because I've got so much in my head at the moment, I don't know how to get anything out.

What I have been doing is a bit of cybernetic evangelism. It's been great to take people to the Stafford Beer archive and watch their heads explode! It's quite a predictable thing...

But I'm also thinking about Beer's holism, and the way that his approach unfolds from nothing. It's this "unfolding from nothing" which really fascinates me. It's rather like my friend Peter Rowlands's work on physics which expands from the idea of nothing (or his 'nilpotent' formulae). Is everything really nothing? What's real about nothing?

I have some good friends visiting me at the end of the week who also know about this stuff. We're going to do a session in the Beer archive. I'm hoping some colleagues from Liverpool also come along.

It's weird how things start to fit together. I suppose my biggest interest in cybernetics at the moment is coherence: how do things fit together? The mechanistic/stochastic approach of cybernetics doesn't address this question well. But David Bohm's idea of the hologram (which Beer was at least aware of because he had a book on it), does.

All the things that fascinate me most, such as music, or conversation, or narrative, or biological form, are all coherent. Are they all holographic?

This question about holograms - particularly in music - has taken on a new dimension for me. I am also working with machine learning tools for a big medical project. We have a problem: how to adjust the judgement of a trained network without screwing up all the other judgements of the network. It's not really how machine learning is meant to work.

We have a brute-force fix for the problem, but it's rather unsatisfactory (and probably not reliable). It would be much better to have a better understanding of what is happening in a convolutional neural network (CNN)

It turns out that current thinking is that it is a kind of hologram. CNNs encode differences and orders of things in a fractal structure. That means that the nature of the problem we want to solve is "where to change the fractal/hologram so that a specific change in its ordering may be made".

This is rather like asking "where to change the  fractal holographic image so that the hologram changes shape in a particular way". The nature of the challenge can be seen by examining a holographic plate:

What it encodes is the interference pattern of light. That means actually that it encodes distance and time (time because interference involves frequency). Now if we can understand how the interference pattern arises, we might be able to understand where to manipulate it. 

Music, I suspect has a holographic structure. I'm writing a paper at the moment where this structure is considered using the fractal images of anticipatory systems which were developed by Daniel Dubois. His images look like this:
In music's hologram, I think what is encoded is the interference between different redundancies. Like light, this interference means that the hologram encodes time and difference. Because we can play with music in a very practical way, maybe there is a way in which insight gained from music can help with the practical problem of machine learning. 

Let's see. 

In the meantime, I've felt the need to shake up my metasystem...

Thursday, 15 November 2018

Stafford Beer's Critical Holism in Education

I gave a presentation about how Stafford Beer's work relates to education to a small group of people from the education faculty at Cambridge last week. I wanted to avoid presenting Beer's work as a kind of fait-accompli, where the Viable System Model (VSM), or Syntegration is the answer (I think this kind of evangelism is very off-putting). But his work is mind-blowing, and if he didn't "have the answer", he certainly had an important way of asking practical questions which is sorely missing from anything in the educational discourse today.

The problems - the reasons why the VSM or Syntegration isn't the answer - or indeed, any other cybernetic theory cannot provide a full answer - are that fundamental problems of time, meaning, emergence, non-ergodicity and coherence haven't been resolved in any of the systems sciences. This is why, for example, the question of agency in cybernetic descriptions is such a problematic question: "where's the person? They're in the recursions", which leads to a slight air of dissatisfaction. We can work to improve this situation - but this will only happen with a critical engagement with cybernetics.

This is not to take anything away from Beer. He nailed what he was doing and what cybernetics is really about: "Cybernetics is about holism". Yes. There are of course many many definitions of cybernetics, which describe it as "ways of thinking", or "ways of thinking about ways of thinking", "the art and science of defensible metaphors" (!), or "the science of effective organisation" - it all gets rather philosophical, giving a newcomer the feeling that they've arrived in some kind of cult. But, in the end, what unites them all is that they all deal with wholes. They all run counter to reductionism.

Holism has a bad name. It is rather closely associated with cults, with theories of everything. But this isn't what Beer meant. He was after (and indeed possessed) a science of holism (notwithstanding the problems raised above). If it is wholes we have to grapple with, and not parts, then we need to know how wholes work - and they are not simple things, but once opened out, they reveal a structure. It is this structure which can be studied and experimented with.

The structure unfolds because whatever whole is considered contains things which cannot be decided. I have recently preferred simply to talk about uncertainty. The point is that this uncertainty has to be dealt with, and by definition, it cannot be dealt with within the "whole". So any whole requires a metasystem - something which sits outside the whole and mops up the uncertainty. It does it, often, by imposing categories for dealing with the uncertainty. It's the metasystem where the reductionism goes on!

Beer knew that there were good and bad ways in which the relationship between a whole and a metasystem could work. If education is seen to be a "whole", then the metasystem has to mop up things like uncertainties over teacher and student "performance": it invents categories and metrics to measure teaching and learning. It even ties some of these metrics to the pay or job security of teachers. More recently it deploys technologies to reinforce these metrics. What happens? "explosive complexification".

Why do these uncertainties arise in the first place? What is it about the whole which invites pathological metasystemic regulation? There's a simple answer to this. It is the hierarchical structures of organisation which education adopts. These structures themselves are very poor at mopping up their own uncertainty: hierarchies attenuate complexity from their bottom to their top, and from the environment to each individual. The only mechanism they have for managing uncertainty is authoritarianism, and this eventually leads to collapse.

What is required are forms of organisation which manage their uncertainty effectively. In education, the most effective way any individual - whether teacher or learner - can manage their uncertainty is to talk to others: "What do you think?" The best form of educational organisation is one which creates the conditions for conversation. Here, Beer's holism suggests that the way to do this is to disrupt the metasystems of each individual. This is really what he attempted with his Syntegration technique. It's what Von Foerster articulated when he spoke about education's role in learning to ask "legitimate questions", or questions to which nobody knows the answer:

  1. “Education is neither a right nor a privilege: it is a necessity.” 
  1. “Education is learning to ask legitimate questions.” 

A society who has made these two discoveries will ultimately be able to discover the third and most utopian one:
  1. “A is better off when B is better off.” (Von Foerster, Understanding Understanding, p209)
Understanding how Von Foerster gets from 2 to 3 is core to appreciating the power of Beer's Critical Holism.