controls – Page 14 – Maverisk / Étoiles du Nord

~vergent predictions, Do or Don’t

This idea, or lack of it, crossed my mind:
When it comes to predictions, following the lead of Tetlock’s Superforecasters may very well work (though note much of it starts with the, sort-of, mental, 50-50 approach of soberly realizing that one may improve, by admitting imprecision and those that claim precision or high scoring rates are wrong) … for issues and questions that converge on one, somewhat exactly determinable, outcome. This, all being within the realm of said book which is very much recommended by the way.
Where some questions, like “What is the best strategy?” may not have such a single outcome; the world changes, and (business-like) having a vision is a grand prediction already. Let alone that the ‘mission’, one’s desired place in that vision of how the world will be in the future, (often / always without a miss) skips the implicit choice issue of what one’s future place could be within that, vaguely defined, future state of affairs. Even if you shoot for the moon [and end up in an infinite and infinitely cold vacuum, among the stars but near-infinitely dwarfed by them] and miss, you may end up in a not-first but still pretty comfortable position; no hard feelings. … This, as an explication of what I’d call diverging predictions: Wide-ranging future states that you might ‘predict’ but most probably in a vocabulaire that will not be valid or understood in the future so traceability of your predictions is … quite close to zero hence your advance predictions have no worth ..! This of course is also in the book but still, too often not realised.

Now, let’s combine this with Maister’s Advisor let alone simple consultancy …

Oh well. Plus:

[Predicting quality of resulting still wines … for second fermentation, mariage, and onwards — priceless; Ployez-Jacquemart]

Right. Explain.

Well, well, there we were, having almost swallowed all of the new EU General Data Protection Regulation to the … hardly letter, yet, and seeing that there’s still much interpretation as to how the principles will play out let alone the long-term (I mean, you’re capable of discussing 10+ years ahead, aren’t you or take a walk on the wild side), and then there’s this:

Late last week, though, academic researchers laid out some potentially exciting news when it comes to algorithmic transparency: citizens of EU member states might soon have a way to demand explanations of the decisions algorithms about them. … In a new paper, sexily titled “EU regulations on algorithmic decision-making and a ‘right to explanation,’” Bryce Goodman of the Oxford Internet Institute and Seth Flaxman at Oxford’s Department of Statistics explain how a couple of subsections of the new law, which govern computer programs making decisions on their own, could create this new right. … These sections of the GDPR do a couple of things: they ban decisions “based solely on automated processing, including profiling, which produces an adverse legal effect concerning the data subject or significantly affects him or her.” In other words, algorithms and other programs aren’t allowed to make negative decisions about people on their own.

The notice article being here, the original being tucked away here.
Including the serious, as yet very serious, caveats. But also offering glimpses of a better future (contra the title and some parts of the content of this). So, let’s all start the lobbies, there and elsewhere. And:

[The classical way to protect one’s independence and privvecy; Muiderslot]

Local jargon

… Suddenly it struck me. In my usual rants against ‘governance’ as in many of my earlier blog posts, the non-existent airheaded formalities that stand in the way of real, not the deflated style of management, I forgot one piece:
The phrase makes sense only in (totalitarian [here I go again], calcifying) bureaucracies. There, the shuffling of empty labels has replaced actual management and ‘governance’ may be used as a placeholder (as said: empty, otherwise the label doesn’t apply fully) for the ‘management’ pastiche that is expected. Outside of those dinosaur organisations (heh, see this), no-one has any need, or place, or (need for) understanding of ‘governance’ and anything sycophanting towards it, will fail to achieve anything close to a positive contribution (though negative contributions, in stifle, overhead, disturbance of good business, may be wide spread). See a business/organisation dabbling in ‘governance’ babble, and you see failure ahead.

I’ll leave it there for you. With:

[For no apparent reason, a whopping crazy car park; Amsterdam]

Walnuts, brain size and you

Combining some recent news, some really old news, and your place in between. Or not.

The recent news: Birds might have tiny brains, but they still may be very intelligent (as animals go). Now, on a related note, discoveries show that the brain cells of birds may be smaller and/or much denser packed than they are in, e.g., humans and family.
The really before-stone-age news: dinosaurs-picture-is-bleak

Combined: Birds have a separate line of descendance from their dinosaur-time quite-close equivalents. Having survived some dino extinction rounds and still remain quite similar in body and operations as before, having kept the same lightweight and small-package brain structure too?
Then, maybe the dinosaurs weren’t so stupid either with their small but possibly also very densely packed neurons and they just had a bad hair day (that’s what you get when a comet strikes your coiffure — footballers beware).
Just a, very,very,very after-the-facts hypothesis… And:

[For wine making; isn’t that obvious !?!?!? Quinta do Vallado; Douro]

The Learning-from-error Error

[Thread development; under ~ ]

Tell me, did you go to school somewhere? Did you finish it, and/or completed assignments and exams to somewhat satisfactory degrees?

Congratulations… To the ‘common’ wisdom that one only learns from error, you have failed. In life.

Because, according to too many, fail fast fail often is the best way to gain knowledge about what doesn’t work — automatically leading to the assurance that doing things differently, will work. If you tried and the result didn’t fail, you haven’t learned. So, if you just learned what centuries of the most learned men (plus women…) brought to you, and achieved, acquired any compound body of knowledge, you may have knowledge but are useless otherwise, like an encyclopedia without a reader? Like some millennial that can google anything but doesn’t know (sic) how to apply the search results (let alone qualify them in the tremendous bias that’s in there)? Or did you learn about process and application along the way …?

Thus, all that human culture is; transferred knowledge on facts, process and application, is denied. Where even Neanderthals had culture and knew how to learn from what had — positively — shown to work in practice (i.e., application, intelligence), you the fail-oriented stumbler, don’t reach up to their level of survivability.

Which leads to both this and this, with a large dose of this. ‘Traditional’ learning, and building enterprises that can last for centuries (or, until the wisdom is lost due to ‘CEOs’ and ‘managers’ quae non), as an antidote and sensible path.

Now, if you can just leave us sanes, the rest of the world to actually be successful in the long and short runs …? Plus:

[Just two boats, or an Atlantic Ocean of knowledge ..? Off Foz]

Another Q

Yet another, relatively (sic) random, quote with a kicker in the tail:

In support of this distinction, Chalmers introduces a thought experiment involving what he calls zombies. A zombie is an entity that acts just like a person but simply does not have subjective experience — that is, a zombie is not conscious. Chalmers argues that since we can conceive of zombies, they are at least logically possible. If you were at a cocktail party and there were both “normal” humans and zombies, how would you tell the difference? Perhaps this sounds like a cocktail party you have attended.

Again, from Ray Kurtzweil’s How to Create a Mind (p.202).
And, of course:

[Just like that; Aachen]

Not just Q, IQ

Well, yesterday’s post was about just a quote, this one’s about what should be a full cross-post but hey, I’m no wizard I’ll just blockquote it from here because it’s so good (again, qua author):

Society in the Loop Artificial Intelligence

Jun 23, 2016 – 20:37 UTC

Iyad Rahwan was the first person I heard use the term society-in-the-loop machine learning. He was describing his work which was just published in Science, on polling the public through an online test to find out how they felt about various decisions people would want a self-driving car to make – a modern version of what philosophers call “The Trolley Problem.” The idea was that by understanding the priorities and values of the public, we could train machines to behave in ways that the society would consider ethical. We might also make a system to allow people to interact with the Artificial Intelligence (AI) and test the ethics by asking questions or watching it behave.

Society-in-the-loop is a scaled up version of human-in-the-loop machine learning – something that Karthik Dinakar at the Media Lab has been working on and is emerging as an important part of AI research.

Typically, machines are “trained” by AI engineers using huge amounts of data. The engineers tweak what data is used, how it’s weighted, the type of learning algorithm used and a variety of parameters to try to create a model that is accurate and efficient and making the right decisions and providing accurate insights. One of the problems is that because AI, or more specifically, machine learning is still very difficult to do, the people who are training the machines are usually not domain experts. The training is done by machine learning experts and the completed model after the machine is trained is often tested by experts. A significant problem is that any biases or errors in the data will create models that reflect those biases and errors. An example of this would be data from regions that allow stop and frisk – obviously targeted communities will appear to have more crime.

Human-in-the-loop machine learning is work that is trying to create systems to either allow domain experts to do the training or at least be involved in the training by creating machines that learn through interactions with experts. At the heart of human-in-the-loop computation is the idea of building models not just from data, but also from the human perspective of the data. Karthik calls this process ‘lensing’, of extracting the human perspective or lens of a domain expert and fit it to algorithms that learn from both the data and the extracted lens, all during training time. We believe this has implications for making tools for probabilistic programming and for the democratization of machine learning.

At a recent meeting with philosophers, clergy and AI and technology experts, we discussed the possibility of machines taking over the job of judges. We have evidence that machines can make very accurate assessments of things that involve data and it’s quite reasonable to assume that decisions that judges make such as bail amounts or parole could be done much more accurately by machines than by humans. In addition, there is research that shows expert humans are not very good set setting bail or granting parole appropriately. Whether you get a hearing by the parole board before or after their lunch has a significant effect on the outcome, for instance.

In the discussion, some of us proposed the idea of replacing judges for certain kinds of decisions, bail and parole as examples, with machines. The philosopher and several clergy explained that while it might feel right from a utilitarian perspective, that for society, it was important that the judges were human – it was even more important than getting the “correct” answer. Putting aside the argument about whether we should be solving for utility or not, having the buy-in of the public would be important for the acceptance of any machine learning system and it would be essential to address this perspective.

There are two ways that we could address this concern. One way would be to put a “human in the loop” and use machines to assist or extend the capacity of the human judges. It is possible that this would work. On the other hand, experiences in several other fields such as medicine or flying airplanes have shown evidence that humans may overrule machines with the wrong decision enough that it would make sense to prevent humans from overruling machines in some cases. It’s also possible that a human would become complacent or conditioned to trust the results and just let the machine run the system.

The second way would be for the machine to be trained by the public – society in the loop – in a way that the people felt that that the machine reliability represented fairly their, mostly likely, diverse set of values. This isn’t unprecedented – in many ways, the ideal government would be one where the people felt sufficiently informed and engaged that they would allow the government to exercise power and believe that it represented them and that they were also ultimately responsible for the actions of the government. Maybe there is way to design a machine that could garner the support and the proxy of the public by being able to be trained by the public and being transparent enough that the public could trust it. Governments deal with competing and conflicting interests as will machines. There are obvious complex obstacles including the fact that unlike traditional software, where the code is like a series of rules, a machine learning model is more like a brain – it’s impossible to look at the bits and understand exactly what it does or would do. There would need to be a way for the public to test and audit the values and behavior of the machines.

If we were able to figure out how to take the input from and then gain the buy-in of the public as the ultimate creator and controller of this machine, it might solve the other side of this judicial problem – the case of a machine made by humans that commits a crime. If, for instance, the public felt that they had sufficient input into and control over the behavior of a self-driving car, could the public also feel that the public, or the government representing the public, was responsible for the behavior and the potential damage caused by a self-driving car, and help us get around the product liability problem that any company developing self-driving cars will face?

How machines will take input from and be audited and controlled by the public, may be one of the most important areas that need to be developed in order to deploy artificial intelligence in decision making that might save lives and advance justice. This will most likely require making the tools of machine learning available to everyone, have a very open and inclusive dialog and redistribute the power that will come from advances in artificial intelligence, not just figure out ways to train it to appear ethical.

Credits

•Iyad Rahwan – The phrase “society in the loop” and many ideas.
•Karthik Dinakar – Teaching me about “human in the loop” machine learning and being my AI tutor and many ideas.
•Andrew McAfee – Citation and thinking on parole boards.
•Natalie Saltiel – Editing.

And, of course for your viewing pleasure:

[Would AI recognise this, an aside in the Carnegie Library; Reims]

Nopsrisk, Irisk

When it’s time, it’s time. Of course, meaning that the tough get going.
Lately, there has been a resurgence in Risk Management. In particular, in Operational risk management. That has been outclassed. Due to, among others, the calimero hanging-on at the tails of financial risk management but having failed to gain traction because the latter’s models were wholly inapplicable and seriously outright unusable for ops risk, due to having no clothes of one’s own (still, the upstart little peasant kid wanted to be emperor), due to having been outflanked by its little nephew of Information / IT Risk Management. That took on the coat of ‘cyber’ (#ditchcyber!) and gained prominence on all the vast wastelands that were left for the picking — and are now overwhelming the heartland with their successes in actual, frontline, FLOT hand-to-hand combat and battles (won).

Time, maybe, to give IRM the prominence it deserves, and forego the subsumption under ops risk ..?

It’s nothing personal…

[Soon again: Serralves]

Not your clients!

An outcry: Stop calling ‘clients’ what are just mass tools to make a profit (incl public sector…) for your actual clients…!

When, why, did the non-politically grossly in-correct usage of ‘clients’ come from, where not only the Facebooks of this world will serve you crumbs and deliver your value to others ..? Because all sorts, yes the dullest of dullest too or in particular, of public sector organisations fall prey to the emptiest of sympathies when they denote their fully captives as ‘clients’, or at best, ‘civilians’ as if they themselves are not the most average, mediocre, irrelevant of those denominiations themselves ..? ‘Clients’ of a social services organisation ARE NOT; apologies for the shout, they are captives, with no alternative to turn to (like actual clients could) but the actual client is some politician(s) that have just enough brains to be the last one standing / clinging to their seats while everyone of anything approaching intelligence even at great distance, will have left or have been pushed out by actually caring for the ‘clients’s interests.
‘Clients’ are just the mass fodder, nothing (sic) more despite all the efforts to paint a social, relating picture.
Get real. Stop the outright lying.

Oh well.

[Actual palace of the People; of course this is Pistoia]

DAUSA

Maybe we should just push for a swift implementation of the megasystem that will be the Digitally Autonomous USA. No more need for things like a ‘POTUS’, or ‘Congress’ or so. When we already have such fine quality of both and renewal on the way into perfection (right?), and things like personal independence and privacy are a sham anyway, the alternative isn’t even that crazy.

But then, there’s a risk (really?): Not all the world conforms yet to, is yet within, the DAUSA remit. Though geographical mapping starts to make less and less sense, there’s hold-outs (hence: everywhere) that resist even when that is futile. The Galactic Empire hasn’t convinced all to drop ~~the Force~~ irrationality and take the blue pill, though even Elon Musk is suspected of being an alien who warns us we’re living in a mind fantasy [this, true, actually — the story not the content so much].
But do you hope for a Sarah Connor ..? Irrationality again, paining yourself with such pipe dreams.

On the other hand … Fearing the Big Boss seems to be a deep brain psychology trick, sublimating the fear of large predators from the times immemorial (in this case: apparently not) when ‘we’ (huh, maybe you, by the looks of your character and ethics) roamed the plains as hunter-gatherers. So if we drop the fear, we can ‘live’ happily ever after; once the perfect bureaucracy has been established. Which might be quite some time from now you’d say, given the dismal idio…cracy of today’s societal Control, or may be soon, when ASI improves that in a blink, to 100,0% satisfaction. Tons of Kafka’s Prozesses be damned.

Wrapping up, hence, with the always good advice to live fearlessly ..! 😉

[Some Door of Perception! (and entry); De Haar castle]