Monday, April 3, 2017

Understanding "Talking Lions" and Other "Black Boxes"

If a lion could speak, we could not understand him.
-- Wittgenstein[1]

Artificial intelligence is everywhere. But before scientists trust it, they first need to understand how machines learn.
-- D Castelvecchi, Can We Open the Black Box of AI? [2]

"With artificial intelligence, we are summoning the demon."
-- Elon Musk (2014) Reported in CNN Tech , Oct 28th.

Faust with Homunculus

Black Boxes Abound. Less so, our understandings of them. The "black boxes" are not only lions, or AI devices, but also humans, and even more. If we generalize, keeping in consideration our perceptual limitations, any source of emissions that appear to us to be more than random, from any object or even an "empty" focus of attention, is, initially, a black box.

Many commentators have remarked that Wittgenstein makes an incautious jump from speaking to understanding in the quote cited above. I will generalize their concerns: how do we know that a (what appears to us to be) non-random emission of, say, sound, is speech? Or, if the transmission is not sonic to humans, language? Animals can learn to mimic human emissions. Even non-English recordings, played backwards, can sound like English speech.

To many a monolingual English speaker, the following phrases sound like pieces of English nursery rhymes colored with a non-English accent:
1. French - Un petit d'un petit s'étonnent aux Halles (Humpty Dumpty sat on a wall.)
2. German - Oh wer, oh wer ist Mai Lido doch Gong? (Oh where, oh where is my little dog gone?)
3. Spanish (Caribbean) -- Grima! Sí, comí. Te excusé que rifa. (Christmas is coming, the goose is getting fat.) [3]
Black Boxes are "Understandings." How does Wittgenstein even judge what the lion is doing? What assumptions is he making? Even if the lion's emissions sound like an English sentence, what is relevant in the context? What is the lion doing? Suppose, for example, our friend, Harry, is standing before a lion's cage in which a cave has been constructed as a den. The lion, staring at Harry, emits the sound sequence, /2aym+ gowing+3 hówm1/ [4]

It sounds like the lion just said, "I'm going home." But did it "say" that? Suppose the lion then lies down and rolls over on its back. It starts to snore. Are we still inclined to think that the lion talked? (If so, did he talk to Harry? Did the lion inform Harry it was "going home?" And, thus, was the lion telling a lie?) Can Harry believe his eyes and ears? Has he jumped to conclusions? (See Artificial Intelligence Weirdness. Need categorizing relate to visual cues?)

The basic problem is how we distinguish illusions, visual, as well as audial or other, from realities. Even AI's have this problem.(See Autonomous Car Collides with Bus: an illusion of abstractions?)

And What Are Understandings? Understandings, when articulated in language, are narratives (or, if highly structured, a program) of collection or connection. We may judge them incorrect, false, or incomplete but they are still understandings -- we might say, "misunderstandings."

People may not be at all articulate as to what they understand so we may have to observe how they proceed from their present conditions toward those outcomes that we judge them to be pursuing. If we observe persistent failure to achieve a goal we might reasonably judge that their understanding of how to achieve that goal is deficient.

There is a large number of likely misunderstandings that persist in any population because they need not or cannot often be put to any test whose outcomes enjoy broad consensus as to their pertinence. (So what if kids believe in the Tooth Fairy!) These questionable understandings are often characterized as (empirically) "non-disconfirmable" beliefs. Examples are:
a. The mind is co-extensive with the body, or:
b. The mind is not coextensive with the body.
c. The universe is a hologram; or
d. The universe is not a hologram.

It would appear that any rationales, or chain of rationales, that contain misunderstandings would thereby be severely weakened. (Many pundits who presume to speak for Science share such misunderstandings with those who presume to speak for Religion. See Pseudo-Science: the reasonable constraints of Empiricism.)

The Fractal(?) Generation of Rationales. Understanding in general is the ability to produce chains of behavior or their narratives which link a confronted situation of interest to a goal to be achieved. But, understandings alone are often narrow and give may give no indication of the interests or abilities of the persons who understand and may be expected to act. Thus, understandings may be just part of what we're looking for.

Rationales can be elaborated from understandings. Rationales often bring up systemic concerns about, say, indicator validity (Cue), actor interests (Concern), and actor abilities (Control) which underly attempts at interventions based on narrower technical understandings.

Understandings can be chained together to produce rationales for action, and rationales themselves chained to produce broader understandings. Thus we may go from merely speculative understanding of how to cross a river, to a rationale for expanding commerce across that river, to an understanding of how political influence can be brought about reiteratively by using market rather than, say, armed forces. Understandings developed this way can be connected together, in chains or trees, etc., to articulate, say, foreign economic development policies. (See The Fractalization of Social Enterprise)

Why Can We Understand Human Black Boxes but find It harder to Understand AI Black Boxes?
It is because many people working in AI feel restricted to technical issues of understanding the how's of economically pertinant AI functioning. Despite the the persistent anthropomorphizing of AI output -- and despite the suspicions that the ultimate goal of AI research is to create some kind of homunculus -- issues of rationale, especially in the contexts of human development and learning are discounted as off target. [5]

Autonomous AI stimulates the same kind of misgivings that one might feel about extra-terrestrials. We are not sure we can predict, much less control what they might do with us. It is a non-sequitur to believe that high computational ability equates to altruism. (See METI. Here We Are! Come Eat Us! Our Children Are Especially Tasty!)

We have built-in, so to speak, Cue-identifying abilities inherited through evolution, both physical, mental and social, that are sensitive to the norms of the environments of our development. We are Concerned to share a world with motile, sometimes dangerous, beings that we can hear, feel, smell, taste as well as see. And we know from personal experience what fear, hunger, danger, hate and social attraction are. [6]

We measure the autonomy we are willing to extend to our ancient non-human friends and enemies because we can guess well what drives them, and how to accommodate them to our society. And, not the least important, we can, with more or less success, Control and defend ourselves against them.

To pursue the issues of understanding and rationale, see (See Intervention. Helping, interfering or just being useless?)

Cordially, EGR

(P.S. Check out this interesting interview with Gary Marcus called Making AI More Human.)

[1] L Wittgenstein 1958b. Philosophical Investigations. ed. G.E.M. Anscombe and R.Rhees. tr. G.E.M. Anscombe, 2nd edition. Oxford: Blackwell. page 223. See comments by Simon van Rysewyk at Wittgenstein Light

[2] D Castelvecchi Can We Open the Black Box of AI? Nature magazine, October 5, 2016

[3] Sources for pseudo-English concoctions: French - Mots d'Heures: Gousses, Rames, The d'Antin Manuscript. Penguin,1980; German - J Hulme Mörder, Guss Reims. The Gustave Leberwurst Manuscript, Clarkson N Potter, 1981; Spanish -- My confabulation, EGR

[4] IPA, digits indicate tone levels. Cf. H A Gleason, An Introduction to Descriptive Linguistics (New York: Holt, Rinehart and Winston, 1955)

[5] See EG Rozycki Behavior in Measurability and Educational Concerns 

[6] See D Gross, Why Artificial Intelligence Needs Some Emotional Intelligence

Thursday, February 16, 2017

Merchandizing Science (and other things...?)

We cannot act on information alone; the information must first be understood, then interpreted for relevance, and finally command belief and commitment. But what if the citizen cannot assess the truth of the available information or its import for action? --- Harry S. Broudy, Truth and Credibility. The citizen's dilemma (New York: Longman, 1981) p. 13.
Information: Food or Sewage? The expression "GIGO," Garbage in, garbage out, used to have some meaning even beyond the world of IT enthusiasts. Now, even though it is more widely recognized, it becomes evermore reduced in cogency. Broudy's comment above was, in 1981, still merely a worry; it was dismissed by many who persisted in letting a smile be their umbrella.

However, Broudy's comment rings true as we see long revered, even though lip-serviced, standards of decency, civility and authority publicly, blatantly disregarded; especially, by self-proclaimed "pragmatic" leaders. The distinction between the concepts of "leader" and "usurper" has become increasingly blurred. (See Leadership as Usurpation)

Foresightfulness, consideration, tolerance, fairness, compassion and, especially, memory are — often publicly and with celebration — cast aside, as "activists" of all persuasions rush to exercise their roles as "change agents" or "disrupters" under the banners of "democracy," "liberty," "Amendment-protection," "entrepreneurship," "greatness" and "social justice."

There seems to be a bias in our culture that anything characterized as research is akin to Godly Revelation. But it's no news that RESEARCH can be badly planned, badly executed, off-target or trivial. And, increasingly, individuals from scientific, religious and governmental communities either rush to participate,or, more passively, support the scramble with contributions that result from their own self-serving miscreation, e.g. tampering with research, faking news reports, ignoring critical violations of law or producing, often dangerous, sub-standard products. (See Retraction Watch)

Why does this happen? It seems that it is because we treat failure as the greatest of disgraces. Individual persons or (usually small) enterprises are expected to take great risk with their resources to provide social benefits without a compensating safety net should they fail. Yet ridicule awaits them as "losers." This anti-failure taboo is so great that even celebrities would rather be caught publicly in a lie about their attempts, than to own up about their failures. (See Barbara Ehrenreich's (2009) Smile or Die ISBN:9781847081735)

Example: The Mysterious "Impossibility" of Multitasking. President Gerald Ford used to be characterized by some as "unable to walk and chew gum at the same time." However, considered judgment held it that the intelligent and accomplished Mr. Ford had been targeted by animadversion. He was not only able to walk and chew gum at the same time, but, like most of us, do many more complex things by multi-tasking.

But -- according to some highly-degreed pundits, we were wrong! Multi-tasking is impossible! RESEARCH proves it! So it is dangerous to multi-task because we are likely to make many more mistakes because we have to switch attention too much to be accurate while we multi-task.

When do we risk such mistakes? When we multi-task! But then, we CAN NEVER make such mistakes, because as RESEARCH shows, multi-tasking is impossible! We CAN'T DO IT! And since it is, and supposedly has ever been impossible, we COULDN'T HAVE DONE IT! So Gerald Ford's detractors, who were no doubt composing their thoughts as they were typing them up while telling them to listeners and judging their reactions were -- in other words -- multi-tasking. (Is there no longer any sensitivity to self-contradiction?!)

There Are Three Problems within this "debate." The first is that to say that we can't multitask assumes there are clear, generally accepted boundaries that distinguish one task from another. That, when humans who use the word, "task," they are invoking the same criteria of identification that researchers do. This is unconvincing because the latter's conclusions about multitasking are so different from what many intelligent practitioners in various fields think about it and DO with it. Multitasking research offers, apparently, not a correction to our language, but a redefinition. The claim that it is impossible to multitask attracts attention, it seems to be an effective marketing tool. But is essentially a bait-and-switch maneuver. (See Low Cost Interventions for a Better World: Reform by Redefinition?)

The second problem is that multitasking research appears to assume that the human brain is something like a computer with only one processor, so that parallel processing, which many computers employ, doesn't exist for humans. But, even though NPR (Think You're Multitasking? Think Again)
may doff its hat to Multi-tasking Impossibility theories, many others have made similar criticisms to the ones here. (Psychology Today (3/30/11) Myth of MultiTasking)

The third problem is that the debate really deals with paying attention to the implementation of goals. Many tasks we undertake may have become for us, to some extent, automatic. This doesn't mean that they are no longer tasks. And of course breakdown in implementation may occur if we have too many things to pay attention to. This realization, once comprehended, does not promise to have much market impact.

Does a Pluralistic Democracy Require "Sleight-of-Tongue" to Persist? Even if the ambiguities of promotional language have some desired market impacts, this in no way guarantees that we will find more people cooperating on the same page of the same agendas. Nor does it promise eventual reduction of conflict or improved leadership. (See Buffering: Enhancing Moral Hazard in Decision-Making?.) It is snake oil for our societal forebodings.

Is there a future for a nation of persons who will not be concerned beyond their own individual (or familial, or tribal) interests? Are we doomed to a perpetual war of every person against each other? Many people seem to think so to judge by their willingness to put some kind of Leviathan in public office.

Nicholas Rescher in Pluralism. Against the Demand for Consensus. (Clarendon Press. 1993, p.180. ) writes
The key consideration for the conduct of interpersonal affairs is that the activities of people can harmonize without their ideas about ends and means being in agreement. (See also, page 7 on coming to terms with realities of individual difference.)
It's easy to say that we have to learn to get along with each other. But will mental analgesics, happy talk, get us to The American Dream? Or can we just work, one day at a time, at living with a lot of fussin', feuding', and fightin' as we stumble toward our Alabaster Cities, toward Making America Great? Let us be patient, think clearly and hope so.

To follow up on these ideas, see It’s Effective? Effective For What? Maintaining working relationships.

--- EGR

Wednesday, December 7, 2016

Need Everyone Agree?: Changing Resisters to Supporters.

Be careful what you wish for, lest it come true. -- Familiar idiom.
You have probably heard someone say something like
Wouldn’t it be great if everyone nation could agree to live in peace? or, less hopefully,
Wouldn’t it be nice of we could all agree on the kind of country we would like to live in?; or, even,
Wouldn’t it be nice if the neighbors could agree not to play loud music after 10 PM?
Are such wishes at best vain hopes? Or are they something that with reasonable effort and careful planning can be achieved?

Here is a thought experiment to try with your friends, family or students:
a. Fill in the blank with a goal, a situation, condition or event, to complete the following sentence:
Wouldn’t it be a better world if everyone could agree on (Goal).

You could substitute other phrases for “a better world” such as “nice,” wonderful,” “a real moral improvement,” or “peachy keen” or “healthier.” Pick something you can get your group to agree on as a substitution.

b. Once you have your statement, e.g. Wouldn’t it be more environmentally sound for people to agree to ban all automobile traffic from inner city commercial areas?, ask your participants to speculate as to who would likely agree to the statement (and why) and who would likely oppose it (and why).

c. Using the chart available at Assessing the Likelihood of Implementing Change , (ALIC) locate your supporters and resisters and the proportions of each you estimate there are. Fill in Line A on the chart.

d. Continue using the above chart to speculate how much you would have to change the percentages of each subgroup to bring them to the likelihood of supporting the change (here, agree with the statement.) Fill in Line B on the ALIC.

e. Considering the Ways to Overcome Resistance suggested by Kotter and Schlesinger in the ALIC, fill in the grid block the likelihood, or your group’s likelihood, of finding the resources to implement these ways, e.g. education & communication, participation & involvement, … etc.

Now consider the following questions:
1. Is the goal well enough defined to be operationalized to avoid a slow shifting (creep) of the target? (See Operationalization.)

2. Consider some of the costs involved in implementing the changes needed; would they be worth the benefits hoped for? Consider who it is who would perceive the achieved goal as a benefit; and, who would think it to be a cost. (You might want to revisit step b. above at this point.)

3. Considering the costs involved in implementing the changes needed, would it be wiser to redefine the Goal so as to shrink the budget needed? (This is a very common practice in all institutions, public and private.)

4. Are the needed ratios of supporters and resisters (see Line B in the ALIC) likely to remain constant through the change process?

To examine these issues further, if you are involved with education, see POLITICS, CONSENSUS AND EDUCATIONAL REFORM

If you are involved with other kinds of organizations, or if you want a more general overview, see Employee Resistance to Organizational Change

--- EGR