Anatomy

posted by Pastorius @ 6:44 PM Comments: 0

Wealth of Papers from Recent Artificial General Intelligence Conference Held on Google Mountain View Campus

Brian Wang presents a nice overview of the recent 4th Conference on AGI, held in the heart of Silicon Valley. This year's AGI conference seems to represent an important evolution in much of the thinking in the AGI field, with a growing depth and sophistication of approach to the problems involved.

To get a better idea of what I am talking about, here are links to most of the papers presented at the conference.

And here are papers from a special workshop on "Self Programming in AGI Systems"

Videos from 3rd Conference on AGI

Artificial general intelligence of human level or higher, would be a radically disruptive technology to modern societies. Along with breakthroughs in scalable robotics, universal nano-assemblers, and a mastery of biological gene expression, a breakthrough in AGI would quickly overhaul most of the bases of modern economics and most other important foundations of everyday life in high tech societies.

Al Fin cognitive scientists have presented many criticisms to mainstream AI approaches -- particularly to the idea that human intelligence can be represented algorithmically. One of the papers presented at this year's AGI conference elaborates on this idea: "Real World Limits to Algorithmic Intelligence"

The biological basis of mathematical competencies is an interesting look by Aaron Sloman at the development of spatial and mathematical concepts in humans. (via Brian Wang) Sloman touches on the idea of the non-verbal or pre-verbal metaphor, an important key to understanding human learning and thought.

Overall, Al Fin cognitive scientists are pleased at the direction the AGI movement is taking, on the basis of the AGI-4 papers they have read, and on the topics covered generally.

There is no doubt a great deal of hidden treasure in the many papers provided at the conference links above. For those who find this sort of thing interesting, enjoy.

posted by Pastorius @ 7:33 PM Comments: 1

Shortly before his death, Marlon Brando was working on a series of instructional videos about acting, to be called “Lying for a Living”. On the surviving footage, Brando can be seen dispensing gnomic advice on his craft to a group of enthusiastic, if somewhat bemused, Hollywood stars, including Leonardo Di Caprio and Sean Penn. Brando also recruited random people from the Los Angeles street and persuaded them to improvise (the footage is said to include a memorable scene featuring two dwarves and a giant Samoan). “If you can lie, you can act,” Brando told Jod Kaftan, a writer for Rolling Stone and one of the few people to have viewed the footage. “Are you good at lying?” asked Kaftan. “Jesus,” said Brando, “I’m fabulous at it.”

Brando was not the first person to note that the line between an artist and a liar is a fine one. If art is a kind of lying, then lying is a form of art, albeit of a lower order—as Oscar Wilde and Mark Twain have observed. Both liars and artists refuse to accept the tyranny of reality. Both carefully craft stories that are worthy of belief—a skill requiring intellectual sophistication, emotional sensitivity and physical self-control (liars are writers and performers of their own work). Such parallels are hardly coincidental, as I discovered while researching my book on lying. Indeed, lying and artistic storytelling spring from a common neurological root—one that is exposed in the cases of psychiatric patients who suffer from a particular kind of impairment.

A case study published in 1985 by Antonio Damasio, a neurologist, tells the story of a middle-aged woman with brain damage caused by a series of strokes. She retained cognitive abilities, including coherent speech, but what she actually said was rather unpredictable. Checking her knowledge of contemporary events, Damasio asked her about the Falklands War. This patient spontaneously described a blissful holiday she had taken in the islands, involving long strolls with her husband and the purchase of local trinkets from a shop. Asked what language was spoken there, she replied, “Falklandese. What else?”

In the language of psychiatry, this woman was ‘confabulating’. Chronic confabulation is a rare type of memory problem that affects a small proportion of brain-damaged people. In the literature it is defined as “the production of fabricated, distorted or misinterpreted memories about oneself or the world, without the conscious intention to deceive”. Whereas amnesiacs make errors of omission—there are gaps in their recollections they find impossible to fill—confabulators make errors of commission: they make things up. Rather than forgetting, they are inventing.

Confabulating patients are nearly always oblivious to their own condition, and will earnestly give absurdly implausible explanations of why they’re in hospital, or talking to a doctor. One patient, asked about his surgical scar, explained that during the second world war he surprised a teenage girl who shot him three times in the head, killing him, only for surgery to bring him back to life. The same patient, when asked about his family, described how at various times they had died in his arms, or had been killed before his eyes. Others tell yet more fantastical tales, about trips to the moon, fighting alongside Alexander in India or seeing Jesus on the Cross. Confabulators aren’t out to deceive. They engage in what Morris Moscovitch, a neuropsychologist, calls “honest lying”. Uncertain, and obscurely distressed by their uncertainty, they are seized by a “compulsion to narrate”: a deep-seated need to shape, order and explain what they do not understand.

As with the woman who told of her holiday in the Falklands, the stories spun by chronic confabulators are conjured up instantaneously—an interlocutor only has to ask a question, or say a particular word, and they’re off, like a jazz saxophonist using a phrase thrown out by his pianist as the start of his solo. A patient might explain to her visiting friend that she’s in hospital because she now works as a psychiatrist, that the man standing next to her (the real doctor) is her assistant, and they are about to visit a patient. Chronic confabulators are often highly inventive at the verbal level, jamming together words in nonsensical but suggestive ways: one patient, when asked what happened to Queen Marie Antoinette of France, answered that she had been “suicided” by her family. In a sense, these patients are like novelists, as described by Henry James: people on whom “nothing is wasted”. Unlike writers, however, they have little or no control over their own material.

Chronic confabulation is usually associated with damage to the brain’s frontal lobes, particularly the region responsible for self-regulation and self-censoring. Of course we all are sensitive to associations—hear the word “scar” and you too might think about war wounds, old movies or tales of near-death experiences. But rarely do we let these random thoughts reach consciousness, and fewer still would ever articulate them. We self-censor for the sake of truth, sense and social appropriateness. Chronic confabulators can’t do this. They randomly combine real memories with stray thoughts, wishes and hopes, and summon up a story from the confusion.

The wider significance of this condition is what it tells us about ourselves. Evidently there is a gushing river of verbal creativity in the normal human mind, from which both artistic invention and lying are drawn. We are born storytellers, spinning narrative out of our experience and imagination, straining against the leash that keeps us tethered to reality. This is a wonderful thing; it is what gives us our ability to conceive of alternative futures and different worlds. And it helps us to understand our own lives through the entertaining stories of others. But it can lead us into trouble, particularly when we try to persuade others that our inventions are real. Most of the time, as our stories bubble up to consciousness, we exercise our cerebral censors, controlling which stories we tell, and to whom. Yet people lie for all sorts of reasons, including the fact that confabulating can be dangerously fun.

During a now-famous libel case in 1996, Jonathan Aitken, a former cabinet minister, recounted a tale to illustrate the horrors he endured after a national newspaper tainted his name. He told of how, on leaving his home in Westminster one morning with his teenage daughter, he found himself ‘stampeded’ by a documentary crew. Upset and scared by the crew’s aggressive behaviour, his daughter burst into tears, he said, and Aitken bundled her into his ministerial car. But as they drove away he realised that they were being followed by the journalists in their van. A hair-raising chase across central London ensued. The journalists were only shaken off when Aitken executed a cunning deception: he stopped at the Spanish embassy and swapped vehicles.

The case, which stretched on for more than two years, involved a series of claims made by the Guardian about Aitken’s relationships with Saudi arms dealers, including meetings he allegedly held with them on a trip to Paris while he was a government minister. What amazed many in hindsight was the sheer superfluity of the lies Aitken told during his testimony. Some were necessary to maintain his original lie, but others were told, it appeared, for the sheer thrill of invention. As Aitken stood at the witness stand and piled lie upon lie—apparently carried away by the improvisatory act of creativity—it’s possible that he felt similar to Brando during one of his performances. Aitken’s case collapsed in June 1997, when the defence finally found indisputable evidence about his Paris trip. Until then, Aitken’s charm, fluency and flair for theatrical displays of sincerity looked as if they might bring him victory. The first big dent in his façade came just days before, when a documentary crew submitted the unedited rushes of their “stampede” encounter with Aitken outside his home. They revealed that not only was Aitken’s daughter not with him that day (when he was indeed doorstepped), but also that the minister had simply got into his car and drove off, with no vehicle in pursuit.

Of course, unlike Aitken, actors, playwrights and novelists are not literally attempting to deceive us, because the rules are laid out in advance: come to the theatre, or open this book, and we’ll lie to you. Perhaps this is why we felt it necessary to invent art in the first place: as a safe space into which our lies can be corralled, and channelled into something socially useful. Given the universal compulsion to tell stories, art is the best way to refine and enjoy the particularly outlandish or insightful ones. But that is not the whole story. The key way in which artistic “lies” differ from normal lies, and from the “honest lying” of chronic confabulators, is that they have a meaning and resonance beyond their creator. The liar lies on behalf of himself; the artist tell lies on behalf of everyone. If writers have a compulsion to narrate, they compel themselves to find insights about the human condition. Mario Vargas Llosa has written that novels “express a curious truth that can only be expressed in a furtive and veiled fashion, masquerading as what it is not”. Art is a lie whose secret ingredient is truth.

posted by Pastorius @ 2:44 PM Comments: 1

Scientists afflict computers with schizophrenia to better understand the human brain

posted by Pastorius @ 1:39 PM Comments: 2

Imagining a World of Total Connectedness, and Its Consequences

Imagine, Michael Chorost proposes, that four police officers on a drug raid are connected mentally in a way that allows them to sense what their colleagues are seeing and feeling. Tony Vittorio, the captain, is in the center room of the three-room drug den.

He can sense that his partner Wilson, in the room on his left, is not feeling danger or arousal and thus has encountered no one. But suddenly Vittorio feels a distant thump on his chest. Sarsen, in the room on the right, has been hit with something, possibly a bullet fired from a gun with a silencer.

Vittorio glimpses a flickering image of a metallic barrel pointed at Sarsen, who is projecting overwhelming shock and alarm. By deducing how far Sarsen might have gone into the room and where the gunman is likely to be standing, Vittorio fires shots into the wall that will, at the very least, distract the gunman and allow Sarsen to shoot back. Sarsen is saved; the gunman is dead.

That scene, from his new book, “World Wide Mind,” is an example of what Mr. Chorost sees as “the coming integration of humanity, machines, and the Internet.” The prediction is conceptually feasible, he tells us, something that technology does not yet permit but that breaks no known physical laws.

Mr. Chorost also wrote “Rebuilt,” about his experience with deafness and his decision to get a cochlear implant in 2001. In that eloquent and thoughtful book, he refers to himself as a cyborg: He has a computer in his skull, which, along with a second implant three years ago, artificially restores his hearing. In “World Wide Mind,” he writes, “My two implants make me irreversibly computational, a living example of the integration of humans and computers.”

He takes off from his own implanted computer to imagine a world where people are connected by them. The implanted computer would work something like his BlackBerry, he explains, in that it would let people “be effortlessly aware of what their friends and colleagues are doing.” It would let each person know what the others “are seeing and feeling, thus enabling much richer forms of communication.”

Cool. Maybe. But beginning with privacy issues, the hazards are almost countless.

In discussing one of them, he cites the work of Dr. John Ratey, a professor of psychiatry at Harvard who believes people can be physically addicted to e-mail. “Each e-mail you open gives you a little hit of dopamine,” Mr. Chorost writes, “which you associate with satiety. But it’s just a little hit. The effect wears off quickly, leaving you wanting another hit.”

Dr. Ratey, he says, calls this “acquired attention deficit disorder.” Think about how this addiction to the quick informational hit would be compounded many times over by those implanted BlackBerrys shooting off constant information. “The effort would be so low, the rewards so intermittent, and the payoff so good, that a savage compulsion would result.”

Recognizing these dangers, and his own isolation, Mr. Chorost set out to make human contact. About to turn 40, in 2005, he had never been in love. Obsessed with the fact that he is short and deaf, by his own description, he undermined relationships. He enrolled in a workshop that a friend told him was about “love, sexuality and intimacy.”

The workshop was such a success that he attended six more and then became an assistant. Passages describing workshop experiences alternate with erudite passages about technology. Sometimes it’s hard to see the connection. Often it’s cringe-inducing.

Halfway into the weekend, clothes come off and participants are urged to hug someone: “I peered around, looking for likely candidates. I met the eyes of a chubby woman about my height, clothed only from the waist down. She smiled warmly and held out her arms.”

A fascinating discussion of optogenetics research is followed by the story of how Mr. Chorost met the woman who is now his wife, Victoria, on a dating Web site. I’m happy for him — Victoria sounds like a wonderful person. But as in “Rebuilt,” he’s shared too much intimacy, too many confessions.


Avert your eyes and get back to technology. Mr. Chorost’s curiosity is contagious. Even if you don’t quite follow the explanation and graphics about how the brain generates speech (discussing the work of computational neuroscientists at Harvard), you may be glad you tried. Edited out of the final book (I read an early galley proof) is the author’s assurance that if you just get through a few technical paragraphs, you’ll understand “how specific memories and perception can be manipulated.”

I didn’t. But I enjoyed the effort. And I liked the author’s belief in my ability to follow what he was saying.
Michael Chorost is not only a clear and concise science writer, but also a visionary. The coming integration of humans and machines may be a bit further off than he thinks, but he convinced me that we will get there someday.

In Ramez Naam’s book More Than Human I learned of an idea that had been proposed by Rodolfo Llinás, a New York University neuroscientist. It was hair-raising. He suggested that engineers could bundle thousands of slender wires into a cable and insert it into the femoral artery in the groin. They would snake the cable through the bloodstream to the brain, as if doing an angiogram. As the cable entered the brain, the wires would spread out so that each one ended up in a capillary. Once put in place, each wire could detect a single neuron’s firing, and change its firing by pulsing a jolt of electricity to it.

Imagine it: a flower blossoming inside the brain, nanometer stalks splitting away from a micrometer stem. Expanding into every available capillary, touching every cubic millimeter of the brain, collecting terabytes of data in every second. By the same token, it could send in terabytes of data every second. It would be the most intimate interface ever invented. If you connected one person’s wired brain to another person’s, you could literally connect them together; they would have a real corpus callosum joining them (albeit with links of radio waves rather than wires.) And if you connected a number of people to each other via the Internet, then you would have a network in which each node was a human brain. The World Wide Web would become the World Wide Mind.

You wouldn’t think there’s room inside your capillaries to insert any kind of wire, but there is. As the image above shows, each nanowire is less than a micron (a millionth of a meter) across—substantially narrower than a capillary. Llinás’s lab has shown that it can be done in principle. They inserted platinum nanowires into the capillaries of tissue samples and detected the activity of neurons lying next to them. Power tends to dissipate rapidly from extremely thin wires, but researchers are trying, with some success, to create wires that can carry the necessary levels of current.[1]

Larger-scale technologies already exist. Doctors can now thread a tube from the groin into the brain to inject anticancer drugs into tumors. These devices, called microcatheters, are thousands of times wider than nanowires, at half a millimeter to a millimeter in diameter. Nonetheless, they show that it’s possible to go deeply into the brain by threading a wire through the bloodstream. In an article on microcatheters the New York Times quoted a doctor as saying, “Technically, I can go anywhere in your brain.”

Of course, anyone can see problems with using large numbers of nanowires in a living brain. How does one guide thousands of wires through tangled kinks of capillaries? (Brain capillaries are as gnarled and twisted as baobab tree branches.) How does one get each one of them to a specific location? What if the wires get tangled? How do you keep them from shorting each other out? What about blood clotting? What if a wire goes through a capillary wall?
But virtually all of these objections were raised against cochlear implants in the 1970s.

posted by Pastorius @ 10:56 AM Comments: 0

The idea that the universe is a giant virtual reality simulation is a well explored theme in science fiction. Films such as The Matrix have used this premise to great effect.

Now a New Zealand scientist is saying that physicists should seriously explore the idea. Brian Whitworth at Massey University says that it is perfectly reasonable to conjecture that "the world is an information simulation running on a three-dimensional space-time screen". Deciding whether or not this is true is a matter for science to resolve.

Assuming Whitworth is serious, what should we make of this idea? He readily admits that this is a weird idea but points out that it is no more strange than many widely held views in physics such as the many worlds interpretation of quantum mechanics, the big bang and Boltzmann brains.

So how would we be able to tell if our universe was a simulation? Whitworth says that if reality was to do something that information processing cannot, then it cannot be virtual. But he falls short of suggesting what this might be.

(As an aside, there are plenty of mathematical algorithms that are incomputable. They are the products of a physical human mind, so if they count as something that information processing could not come up with, Whitworth's idea is already dead in the water.)

Whitworth goes on to suggest various ways in which phenomenon associated with quantum mechanics and relativity can be explained in terms of VR.

He also claims that VR can resolve many of the philosophical questions associated with the Big Bang, such as what caused it and how could it arise when there was no space and time. His answer is that the universe simply booted up although he conveniently ignores all the questions that such a "Big Boot" would raise.

Whether the VR hypothesis is actually testable is a question Whitworth avoids. But without testable predictions about the universe that would distinguish this idea from other theories, the VR hypothesis is pure philosophy.

That's why it is almost certain to be ignored by mainstream physicists. It's not the first idea to suffer this fate - the physicist David Bohm proposed a small modification to quantum mechanics that made no difference to its predictions but ensured that the theory was deterministic.

Most physicists rejected it on the basis of Occam's Razor: that science should strive for the simplest theory that fits all the facts.

My guess is that Whitworth's work will go the same way.

posted by Pastorius @ 3:54 PM Comments: 1

Browse Music by composer, performer, instrument, period or form

FREE CLASSICAL MUSIC DOWNLOADS

GET MUSIC HERE.

posted by Pastorius @ 3:05 PM Comments: 0