Tuesday, March 27, 2018

duplicitous singularity

re-reading the Altered Carbon trilogy, I'm struck by a fundamental philosophical with re-sleeving (upload)  - while the book runs with the paradigm of digital recording of the essence of consciousness and skates over thin ice about embodied intelligence (the meat is just a machine the s/w of a human being runs on), there's a separate idea of continuity; as with perfect forward and backwards secrecy, how is the new running copy anything to do with previous copies, and how do previous copies anticipate the new copy? they don't  - they are copies. so philosophically, they aren't the same person, they are separate instances. but worse, if there's any notion of "eventual consistency" in how anticipation (and memory) overlap and interleave, then this just doesn't work at all. not one bit.

Thursday, March 08, 2018

What isn't AI?

so data science and its typical tools aren't really AI - machine learning, even deep learning, even generative adversarial nets style deep learning still isn't intelligent, though it sure is artificial - its useful, but claiming that a classifier trained on zillions of human-labelled images containing cats and no cats, is recognizing cats is just stupid - a human can see a handful of cats, including cartoons of pink panthers, and lions and tigers and panthers, and then can not only recognize many other types of cats, but even if they lose their sight, might have a pretty good go at telling whether they are holding their moggy or their doggy - how? well, because humans (probably) have a large collection of tools evolved (and trained) in the brain, and the brain is embodied, and so has perception, interaction, motion, sense of things light touch (how furry is that fuzzy looking cat's tail, how does a cat hold themselves when prowling, playing or just chilling... etc

these tools operate at many levels - some may just be context/recall, some may effectively be analogue programmes that model gravity or other physics things (stuff games software writers call "physics models"), and some may very well look like artificial neural nets (things to de-noise complex signals, and to turn moving images (the retina has gaps and doesn't refresh infinitely fast) into representations that let you find objects and name them (that cool cat is a songby squeeze :-) )

there are feedback loops between the "low level" perception stuff and the high level models so the models are (surprise surprise) nature & nurture...but prefacing learning with a model is going to help with unsupervised learning a lot - if we understand 3D, motion, gravity, materials (skin, bone, fur, muscle, fat, what we're made of, and what these other things are made of, wood, grass, mud, water, air etc) etc, then we don't have to see zillions of images of an X, because we generalize from one or two images to lots of views we'd expect an X to look like when its curled up asleep, or jumping up 3 times its height to catch a bird.

So there's a higher level still than al that, as if that wasn't complicated enough for you: humans (most animals) are pretty autonomous - they have goals (tropisms - find food, find partners, survive, enjoy, avoid pain etc), and they have some slightly less obvious tools like curiosity, imagination, creativity - all with a smattering of randomness. These can help seek out diverse input (and create different interactions) so our low level perception & Interaction are constantly refreshed, and our models are updated by challenges (think, scientific method & falsification & parsimony/occam's razor etc).

Then there can be another level still - things like self-awareness, consciousness,, beliefs, neuroses, and even "taste"/aesthetics, and of course, we are social beings, so ethics and theory of mind, and daft stuff like religion and manifest destiny and other collective psychoses. Ghost bugs in the machine.

Monday, January 29, 2018

data science, ml & ai - so, crates

A young robot's primer on the origins of her learning and intelligence.

Data Science mediates between evidence and knowledge. it is as old as humans.

Plato explains the theory of forms through the allegory of the cave, wherein prisoners can only see
the shadows of puppets on a wall, cast by the light from a fire behind them. the prisoners attempt
to infer what is really going on based on what they see.

The analogy beautifully captures the separation between the imperfect nature of
evidence, and the complexity of the model one might form based on said evidence.

Data (from the Latin for "given") is often flawed. our models (forms) are usually limited.
We have developed natural philosophy, and scientific method, to arrive at the best models we can -
e.g. by parsimony (Occam's Razor). however, for millenia, the job of sifting evidence and creating
models was combined. experts carried out (at least) two tasks.

Statistics emerged as a way to summarise and model the data itself, specialising the business of 
understanding the evidence itself, so that the form of models is abstracted away from the details
of the data (to some degree).

Now, we have moved to a world where the formation of models itself is, to some extent, part of this
new science. in varying degrees, the mechanisation of this process of model formation has moved on.

Machine learning packages up the business of spotting
which puppet is which, and even what the plot is.

Artificial Intelligence attempts to move this on to
figuring out it is a puppet show, inferring that there are puppet masters, 
and deciding what they intend the shadow play to mean.


In the realm of science fiction (The Matrix, Dark City, etc), there are actual puppet masters. In
the realm of superstition, these masters may even be adversarial. Post enlightenment, we usually
assume the puppet masters have no intent, but merely represent a more complex world, for which the
puppets and their shadows give us a chance to understand. Popper's model of objective knowledge
suggests that scientific method allows us to ever improve our (approximate understanding.

The mechanisation of this process starts with the problem of industrial scale data gathering,hence
with industrial scale organisation of society. Hence statistics for things like insurance (actuarial tables)
had to be computers (by people initially; and many other planning tools.

With the emergence of computers & the digital age, we gather more data, store it and can process it
more readily. Excitement over "thinking machines" led to the notion that human intelligence could
be understood through computational models. Early on, Artificial Intelligence researchers realised
that a number of sub-challenges exist, starting from understanding more complex evidence from the
senses - hence image processing and understanding objects and scenes was important from the get-go, for a
robot, for example, safely to navigate its environment. Social intelligence requires communication,
and humans use (largely) speech, so natural language processing was a requirement. Other, grander
challenges concerning consciousness and imagination were envisaged, but there was enough to be
getting on with. Indeed, too much at the time. It wasn't until recent decades that the tools and
techniques scaled up to tackle these problems with comparable efficacy to humans. 

Between the 19th & 20th century advances in statistics, and the discipline of Machine Learning,
there are a great many steps. We are not just concerned with quantities of data ("big data") but
with quality and structure of said givens. There's a huge range in the effort put in to make sure
there are consistent records kept in the simplest system. What those records represent is
also drawn from a wide range of complex structures (including no structure!). The advances in
mathematics and statistics are now coupled with advances in computing platforms to allow
exploration and comprehension of these properties of the data. As we go further down this road, we
start to infer what the data tells us about the world it comes from. This is not easy, as Plato's
allegory illustrates.This is also where we move from statistics to data science - we now have a
discipline which makes (falsifiable) predictions about the shadows and the puppets. Others can see
if out predictions remain accurate, or deviate from what they observe. We can use these predictions
to help look for explanations of what is going on, and choose between different explanations. That
process itself is to an increasing extent also being mechanised (automated). 

However, the biggest advances are really often still the old techniques running on affordable,
very fast hardware with a lot of storage. Rarely do these techniques surprise with something that 
resembles general intelligence. That said, the tools are clearly increasingly very very useful. 
Domains where we have potential to have a huge impact are added every day, and as we gain the
capability to apply data science on new combinations of evidence, the unpredicted or unexpected 
will no doubt emerge. While we started this note with a discussion of the emergence of a science
from the shadows, many of the key applications of the science concern data about us - humans. As we
make more observations, and automate processing and decision making concerning humans, we mustn't
lose sight of that humanity. After all, we are not the puppets.

As Conan Doyle said thru his memorable character Sherlock Holmes: 
"Once you eliminate the impossible, whatever remains, 
no matter how improbable, must be the truth."

---------part two--------

statistics aren't what they used to be....before and after Bayes...and before and after Hinton....

two advances in statistics combine with lots of fast computers and data to yield a lot of the machine learning and AI excitement - that's the way that we update our knowledge when we learn new evidence, due to the reverend Bayes, and the way we avoid having very much explicit domain knowledge in our machine for learning, due to artificial neural networks and "deep" learning 

first off, we need a bunch of data - and typically, we need some clever people to label that data. so now we need to talk about domain knowledge and expertise, and how machine learning and AI has so often been parasitically dependent on human intelligence. this is where we "crowdsource" (or already have lying around) a bunch of data that has aready been classified in some senses as "A" or "not A" - 

so lets work by example: lets say we want to learn a classifier for images with cats. and we have many many images (e.g. from google search or from some other available data) that has been tagged with whether or not there at cats in the image. so now we "fire up" our very simple neural net on this data, to train it in what is "cat-ness"  - input at each step is an image and the label (cat or not cat). the neural net is, effectively, a big table where we write things in on the left hand side columns, and copy values through to the next columns if they fit some value so far (are near enough), and not if they don't. When we get all the way out the far side, we have a column full of numbers and one box that says "is cat" or "isn't cat". we compare that to the input and if we're right, we increase the values in the boxes in the table, and if we're wrong, we subtract from those values. [n.b. this is a gross simplification of how neural nets work missing out hidden layers, convolution, and piles of important other tricks that make things work "better")...] Another way of thinking of the "layers" of the neural net might be like a series of combs, each with different length (adjustable) teeth, which let things flow through to the next comb, and when the answer pops out the end and is right, we leave combs' teeth lengths alone more, and wrong, we adjust their lengths more. sort of:-) The adjustment process uses a technique called stochastic gradient descent - one way to picture that part of the process is like snowboarding many times down a complicated slope, adjusting which way you go on each trip, til you get the best ride. or skateboarding. 

we do that for millions and millions of images. whether the cats are black or pink, furry or bald, the result is a table/box that is really quite good at spotting cats (a trained classifier) - we can then put that table in (say) a camera, and now the camera will be able to say "cat" or "not cat" and tag pictures itself without a human.

we could to the same with (say) machine translation. again, totally dependant on work already done by people, say we have a lot of text that has been translated (e.g. copies of Dickens' novels or travel guides or web pages, hand-translated between French and Latin and Greek and Basque and Korean. We can take the text and pull it into small pieces (words, phrases, sentences) and learn a classifier for these which "recognises" input phrase x in language A and outputs it in B...again, re-enforced over more and more examples, but once built, just a "table" to run things on your laptop...



What happens if we can't get labelled data? at least two reasons - we don't have time, or people just aren't good at this particular task - for example finding stuff in the asteroid belt, or explaining why a set of moves in Go were good (something people can explain in chess:). So one trick is to generate synthetic input to the neural net, where we "know" in the generator whether we put a pink panther in the picture or not. Another might be that there's something implicit in the output of the trained neural net that tells us if it is on the right track or gonig down the garden path - e.g. wins a game or not....so this is where Generative Adversarial Nets come in - where there are now two nets, one is like Socrates or Yoda, and the other is the "student"....

These techniques are very effective nowadays, although one interesting problem is that we're not entirely sure what the "features" are that they react to in all cases. Entirely different approaches to machine learning, involve taking a more explicit collection of rules (perhaps based in a model which we put in, in the first place) and just learning which are more or less important in which combinations - it isn't clear, but some research seems to be suggesting that there's ways to bring these approaches towards each other somehow...


There's much more to be said about other (and sometimes simpler, sometimes more complex) machine learning approaches, and about more foundational work on AI in the sense of trying to do something utterly different from all the above, which is actually understanding how humans (and other smart species) do so much without supervised learning. some the old and new theory for the latter concerns higher level goals & models & intentions....which maybe we just evolved to have, and sit at an entirely different level than the neurones and synapses in the grey matter and why it won't be just something emergent from more silicon....

Thursday, January 25, 2018

Medieval Blockchain

reading david graeber's Debt and learning a lot about what makes the world go round, and it sure isn't "money"

so the cliche narrative for currency starts out using it as a unit of accounting, has it as a mechanism for exchange, and ends up with it storing value...

but before that, it has to kill & eat its ancestors, and it often mis-identifies them as barter, and, perhaps, more correctly, as haggle...

but how would barter actually be a thing? what are the chances, especially in a hunter gatherer community, but even in a simple agrarian society, that I have the very thing you want in the right quantity, and the same time as you have what I want? lets take 150 people and think it through....no, lets not - its just silly. you'd keep some sort of notional tally - indeed, it would be part of your social network - you'd know all those people in your 150 network, how much you "trust" them, and in particular, how much you owe them or they owe you...since when....and whether you ever need to "collect"...

secondly, why would I think something that's worth a handful of fruits to you, is worth a handful of fruits to someone else? more likely, from each according to her means to each according to her needs. obviously - hence, looking at medieval society, a blacksmith would "charge" his lordship hundreds of times the debt that he'd charge his neighbour, the farmer or baker or coalminer, to shoe a horse....the farmer needs the horse to plough the field to grow the wheat to give to the baker to make the bread the blacksmith eats while making horseshoes...the lord only needs shiny horseshoes to look smart at the joust.



lefties ? no, just there's a lot more to how things would work in a truly human economy. a richer world view.

So how would a blockchain support this diversity? certainly not through a blockchain currency. but perhaps through smart contracts, which would unwrap between people, recording, and timeshifting but also value-shifting depending on who the players are on each step of the transaction chain.

a medieval, social-networked based distributed ledger/smart contract/chain space?

Tuesday, January 16, 2018

The Don - Gifts and Obligations


  1. le don - as in mafia - makes you an offer you can't refuse
  2. don - as in oxbridge - works on tech because they can, not because they should
  3. don - the legendary lover - and hub for STDs....

references 

Thursday, December 21, 2017

Fake News is a Thought Experiment

Breaking: There is no fake news - the stories are all part of a thought experiement...

firstly, an experiment on the public, to soften people up to the idea - to see how discourse, thought, in public, changes when doubt is cast on the veracity, the inalienable truth, of every day news - how can anyone campaign to undermine the authority of another, when every so-called fact is called into question.

secondly, an experiment that implies to people that news ever was true. that news channels controlled,, owned, censored, modified, changeable, rarely checked, lacking in provenance, authored, only-too-human, were some actual source of authority (in the sense of expertise or reflecting the disciple and learning of the source and utterance) rather than merely authorship, a form of creative writing...

after all, journalists decode what is happening, then encode it. what is lost in translation, modified in interpretation, distorted through cognitive and perceptive bias, to be refracted through the eyes of the reader."


there is no fake news-  there never was. or there always was and will be. "so high you cant get over it, so low, you can't get under it...

Monday, December 18, 2017

zeroth edition

welcome to the zeroth edition, the world's first bespoke counterfeit book publisher. we specialise in fake books. we work very hard to identify authors who resemble in some fashion or other the original author of an original work. We then commission them to write a book. Unlike George Burger's high end operation with his Front, Peter Lyre, we are not out to achieve better than perfect, but more to democratize the process of authorship, hence our product is often quite significantly different than the 1st edition you may have read at school, or seen dramatized on the BBC. No, we believe in a babel-brook approach, where many many versions flow down to the river of readership, and eventually into the sea of dissatisfied librarians, who now have to decide whether to shelve all the different versions by title, so that A Tail of Twosities is kept close to the authentic A trail of Truieities, and not far from the poor quality work of a similar name, or else to sort the works by author, making it impossible for a lifetime of scholarship to uncover all the versions of so many lives.

If you would like to sign up to be an author, you already have. If you would like to be a reader, you already are. If you are a copyright lawyer seeking fun&profit, you already did.

We realize that there are many other counterfeit book publishers already out there, and why would you come to us rather than them? Indeed, who had this idea first? Well, all we can say is that we are never knowingly underwritten.