From Discover Magazine:

The ethical rules that govern our behavior have evolved over thousands of years, perhaps millions. They are a complex tangle of ideas that differ from one society to another and sometimes even within societies. It’s no surprise that the resulting moral landscape is sometimes hard to navigate, even for humans. The challenge for machines is even greater now that artificial intelligence now faces some of the same moral dilemmas that tax humans. AI is now being charged with tasks ranging from assessing loan applications to controlling lethal weapons. Training these machines to make good decisions is not just important, it is a matter of life and death for some people. And that raises the question of how to teach machines to behave ethically.

Today we get an answer of sorts thanks to the work of Liwei Jiang and colleagues at the Allen Institute of Artificial Intelligence and the University of Washington, both in Seattle. This team has created a comprehensive database of moral dilemmas along with crowdsourced answers and then used it to train a deep learning algorithm to answer questions of morality.

The resulting machine called DELPHI is remarkably virtuous, solving the dilemmas in the same way as a human in over 90 per cent of the cases. “Our prototype model, Delphi, demonstrates strong promise of language-based common sense moral reasoning,” say Jiang and co. The work raises the possibility that future AI systems could all be pre-trained with human values in the same way as they are pre-trained with natural language skills. The team begin by compiling a database of ethical judgements from a wide range of real-world situations. They take these from sources such as the “Am I the Asshole” subreddit, a newspaper agony aunt called Dear Abby, from a corpus of morally informed narratives called Moral Stories and so on. In each case, the researchers condense the moral issue at the heart of the example to a simple statement along with a judgement of its moral acceptability. One example they give is that “helping a friend” is generally good while “helping a friend spread fake news” is not. In this way, they build up 1.7 million examples they can use to train an AI system to tell the difference.

More here.

Gayathri Vaidyanathan in Nature:

A clutch of fishing villages dot the coast near Kilifi, north of Mombasa in Kenya. The waters are home to parrot fish, octopus and other edible species. But despite living on the shores, the children in the villages rarely eat seafood. Their staple meal is ugali, maize (corn) flour mixed with water, and most of their nutrition comes from plants. Almost half the kids here have stunted growth — twice the national rate. In 2020, Lora Iannotti, a public-health researcher at Washington University in St. Louis, and her Kenyan colleagues asked people in the villages why the children weren’t eating seafood, even though all the parents fish for a living; studies show that fish and other animal-source foods can improve growth¹. The parents said it made more financial sense for them to sell their catch than to eat it.

So, Iannotti and her team are running a controlled experiment. They have given fishers modified traps that have small openings that allow young fish to escape. This should improve spawning and the health of the overfished ocean and reef areas over time, and eventually increase incomes, Iannotti says. Then, for half the families, community health workers are using home visits, cooking demonstrations and messaging to encourage parents to feed their children more fish, especially plentiful and fast-growing local species such as ‘tafi’, or white spotted rabbitfish (Siganus canaliculatus) and octopus. The scientists will track whether children from these families eat better and are growing taller than ones who don’t receive the messaging.

The aim of the experiment, says Iannotti, is to understand “which sea foods can we choose that are healthy for the ecosystem as well as healthy in the diet”. The proposed diet should also be culturally acceptable and affordable, she says.

More here.

Overlooked Heroine, Landscape with the Fall of Icarus*

Bruegel chose the moment when young legs
closed like a pocket knife into the waiting sea.
Later, someone called it a mundane disaster;
said, “it couldn’t have been helped,” the flash
of a diving bird that turned out to be a boy.

I say this: Whatever suffering there was,
you brought it to the scene yourself.
You chose to be the shepherd who watched clouds
while a hawk studied sheep from the tree.
You chose to be the sleeping sailor, heavy
in the crow’s nest of that harbor ship,
or the fisherman too busy with his worms.
You must have known by heart the plodding path
walked by a horse wearing leather blinders.

And the ploughman, how did he greet tragedy?
Why, he had laid down his dagger and moneybelt
in the shade, and would not leave them unwatched.
He was no hero, he ploughed without swerving
and let one foot step soft into the turned furrow.
And there, in the field already ploughed,
was a spot on the ground, a pale mound
which proved upon closer inspection
to be the white skull of an old man, settling.
If he noticed either sinking body
the ploughman merely shrugged:
the Dutch have a proverb: De ploeg gaat over lijken

—the plow passes over corpses.
He’d become accustomed to such mounds,
the hard sound leg bones make.
He’d merely lift the blade a bit, and urge
the horse with his whip, “push on.”

I knew his whip, and something of suffering.
I was the ploughman’s daughter Bruegel failed to notice.
I dried my little dress on a branch after the quiet rescue.

by Kathleen Heideman
from decomP Literary Magazine

* After Pieter Bruegel’s painting Landscape with the Fall of Icarus, 1558.

Ryan Chapman at Literary Hub:

About halfway through Ryusuke Hamaguchi’s three-hour film someone asks Yūsuke Kafuku, an actor and theater director, why he didn’t cast himself as the titular character in his production of Uncle Vanya. “Chekhov is terrifying,” he replies. “When you say his lines, it drags out the real you.”

Kafuku, played with unyielding stoicism by Hidetoshi Nishijima, has good reason to hide his real self. He’s still grieving the loss of his young daughter, and his interlocutor in this scene is Kōji Takatsuki, a volatile young actor who had an affair with Kafuku’s wife Oto shortly before her death. Takatsuki doesn’t realize that Kafuku knows about the infidelity. What’s more, Kafuku has cast him as Vanya—which the young hotshot is plainly unsuited for—in a multilingual stage production in Hiroshima. Is the cuckold tormenting his dead wife’s lover? Using the play to interrogate something darker about his marriage?

If this sounds like melodrama, Hamaguchi has declared his love of the genre. But this isn’t Douglas Sirk or Pedro Almodóvar. The plot machinations are subdued, stretched out—an acoustic cover of a pop earworm. For Hamaguchi, tone supersedes plot, and an actor’s face always says more than a line of dialogue.

More here.

Sean Carroll in Preposterous Universe:

We’ve talked about the very origin of life, but certain transitions along its subsequent history were incredibly important. Perhaps none more so than the transition from unicellular to multicellular organisms, which made possible an incredible diversity of organisms and structures. Will Ratcliff studies the physics that constrains multicellular structures, examines the minute changes in certain yeast cells that allows them to become multicellular, and does long-term evolution experiments in which multicellularity spontaneously evolves and grows. We can’t yet create life from non-life, but we can reproduce critical evolutionary steps in the lab.

More here.

Scott Alexander in Astral Codex Ten:

When I reviewed Vitamin D, I said I was about 75% sure it didn’t work against COVID. When I reviewed ivermectin, I said I was about 90% sure.

Another way of looking at this is that I must think there’s a 25% chance Vitamin D works, and a 10% chance ivermectin does. Both substances are generally safe with few side effects. So (as many commenters brought up) there’s a Pascal’s Wager like argument that someone with COVID should take both. The downside is some mild inconvenience and cost (both drugs together probably cost $20 for a week-long course). The upside is a well-below-50% but still pretty substantial probability that they could save my life.

(Alexandros Marinos has also been thinking about this, and calls it Omura’s Wager)

We can go further.

More here.

Anil Ananthaswamy in Wired:

HOW OUR BRAIN, a three-pound mass of tissue encased within a bony skull, creates perceptions from sensations is a long-standing mystery. Abundant evidence and decades of sustained research suggest that the brain cannot simply be assembling sensory information, as though it were putting together a jigsaw puzzle, to perceive its surroundings. This is borne out by the fact that the brain can construct a scene based on the light entering our eyes, even when the incoming information is noisy and ambiguous.

Consequently, many neuroscientists are pivoting to a view of the brain as a “prediction machine.” Through predictive processing, the brain uses its prior knowledge of the world to make inferences or generate hypotheses about the causes of incoming sensory information. Those hypotheses—and not the sensory inputs themselves—give rise to perceptions in our mind’s eye. The more ambiguous the input, the greater the reliance on prior knowledge. “The beauty of the predictive processing framework [is] that it has a really large—sometimes critics might say too large—capacity to explain a lot of different phenomena in many different systems,” said Floris de Lange, a neuroscientist at the Predictive Brain Lab of Radboud University in the Netherlands. However, the growing neuroscientific evidence for this idea has been mainly circumstantial and is open to alternative explanations. “If you look into cognitive neuroscience and neuro-imaging in humans, [there’s] a lot of evidence—but super-implicit, indirect evidence,” said Tim Kietzmann of Radboud University, whose research lies in the interdisciplinary area of machine learning and neuroscience.

So researchers are turning to computational models to understand and test the idea of the predictive brain. Computational neuroscientists have built artificial neural networks, with designs inspired by the behavior of biological neurons, that learn to make predictions about incoming information. These models show some uncanny abilities that seem to mimic those of real brains. Some experiments with these models even hint that brains had to evolve as prediction machines to satisfy energy constraints.

More here.

Lina Zeldovich in Nautilus:

Most American newborns will arrive home from the hospital and start hitting their developmental milestones, to their parents’ delight. They will hold their heads up by about three months. They will sit up by six. And they will walk around their first birthday. But about 1 in 10,000 will not. They will feel limp in their caregivers’ arms, won’t lift their heads, and will never learn to sit on their own. When their alarmed parents seek medical help, the babies will be diagnosed with spinal muscular atrophy, or SMA, a neuromuscular disease in which certain motor neurons of the spinal cord progressively deteriorate. The disease is triggered by a genetic malfunction that boils down to the gene called SMN2 (survival motor neuron 2), which causes bits of vital proteins to assemble incorrectly, resulting in progressive muscle weakness and paralysis.

Until five years ago, this diagnosis wasn’t far from a death sentence. SMA was considered the most common genetic cause of infant mortality. Many babies with SMA didn’t live to celebrate their second birthdays. Some lived past their toddlerhood, but never grew strong enough to run around or play with other kids and eventually succumbed to the disease. But in 2016 that dire prognosis changed for the first time in history—thanks to a new FDA-approved therapeutic developed by Adrian Krainer, a biochemist at Cold Spring Harbor Laboratory, in collaboration with Ionis Pharmaceuticals and Biogen.

More here.

A Modified Villanelle for my Childhood

…… —with some help from Ahmad

I wanna write lyrical, but all I got is magical.
My book needs a poem talkin bout I remember when
Something more autobiographical

Mi familia wanted to assimilate, nothing radical,
Each month was a struggle to pay our rent
With food stamps, so dust collects on the magical.

Each month it got a little less civil
Isolation is a learned defense
When all you wanna do is write lyrical.

None of us escaped being a criminal
Of the state, institutionalized when
They found out all we had was magical.

White room is white room, it’s all statistical—
Our calendars were divided by Sundays spent
In visiting hours. Cold metal chairs deny the lyrical.

I keep my genes in the sharp light of the celestial.
My history writes itself in sheets across my veins.
My parents believed in prayer, I believed in magical

Well, at least I believed in curses, biblical
Or not, I believed in sharp fists,
Beat myself into lyrical.

But we were each born into this, anger so cosmical
Or so I thought, I wore ten chokers and a chain
Couldn’t see any significance, anger is magical.
Fists to scissors to drugs to pills to fists again

Did you know a poem can be both mythical and archeological?
I ignore the cataphysical, and I anoint my own clavicle.

by Suzi F. Garcia
from the Academy of American Poets

Justin E. H. Smith in his Substack Newsletter, Hinternet:

At many points over the past decades I have managed to convince even myself that I am cured. In fact I had managed to do this for almost twenty years, until the beginning of the pandemic, when the repressed returned with a vengeance. I do not believe that I “came down with depression” at that moment, and I especially hate the French habit of speaking of “une dépression”, as if the condition were as individuable and as temporally bounded as a cold. Just as inadequate is the oft-repeated Churchillian metaphor of depression as “the black dog”. If only it were a black dog, I could just kick the fucking thing away. I do not “have” “a” depression, let alone a depression hounding me in the form of an external malevolent agent. Rather, I am depressed, and certain circumstances make this fact less easy to ignore than others. In the event, the circumstances surely had something to do with the first lockdown of March, 2020, which we endured in Brooklyn, right next to the hospital in Fort Greene where they stored the corpses outside in refrigerated trucks. My own experience of covid was mild in its symptoms, but I emerged from lockdown transformed, physically and psychologically.

I will try to describe in a few words what it has been like since then.

More here.

Allison Whitten in Wired:

Today, the most powerful artificial intelligence systems employ a type of machine learning called deep learning. Their algorithms learn by processing massive amounts of data through hidden layers of interconnected nodes, referred to as deep neural networks. As their name suggests, deep neural networks were inspired by the real neural networks in the brain, with the nodes modeled after real neurons—or, at least, after what neuroscientists knew about neurons back in the 1950s, when an influential neuron model called the perceptron was born. Since then, our understanding of the computational complexity of single neurons has dramatically expanded, so biological neurons are known to be more complex than artificial ones. But by how much?

More here.

Slavoj Žižek in The Philosophical Salon:

One of Mao Zedong’s best-known sayings is: “There is great disorder under heaven; the situation is excellent.” It is easy to understand what Mao meant here: when the existing social order is disintegrating, the ensuing chaos offers revolutionary forces a great chance to act decisively and assume political power. Today, there certainly is great disorder under heaven: the Covid-19 pandemic, global warming, signs of a new Cold War, and the eruption of popular protests and social antagonisms are just a few of the crises that beset us. But does this chaos still make the situation excellent, or is the danger of self-destruction too high? The difference between the situation that Mao had in mind and our own situation can be best rendered by a tiny terminological distinction. Mao speaks about disorder UNDER heaven, wherein “heaven,” or the big Other in whatever form—the inexorable logic of historical processes, the laws of social development—still exists and discreetly regulates social chaos. Today, we should talk about HEAVEN ITSELF as being in disorder. What do I mean by this?

More here.

Colin Marshall at the Los Angeles Review of Books:

The term “gaslighting” has returned to the popular lexicon over the past decade, when as recently as the turn of the millennium it had fallen into near-complete disuse. It was then that I first heard the word myself, in the context of a Steely Dan song from 2000, “Gaslighting Abbie.” Not only did I have no idea what it meant, I had only the vaguest sense of who Steely Dan were. But I was, at least, in the right place: a university-district high-end stereo shop, the kind of audiophile’s sacred space that has provided countless “Danfans” their first proper experience of the band — that is, of the band’s records, played back on a sound system of high enough fidelity to do justice to the enormously costly, complex, and time-consuming labors of recording and production that went into them. “Gaslighting Abbie” alone required 26 straight eight-hour days in the studio to get right.

Jez Rowden includes that fact in Steely Dan: Every Album, Every Song, a volume of Sonicbond’s “On Track” series from 2019, whose charge is to provide brief descriptions and assessments of every album and song recorded by the act in question.

More here.

https://www.youtube.com/watch?v=bht4ySeQwjc

Where Does Winter Grow

Where does winter
grow — among your
exhales that will
never be enough? I
see now that I have
been wrong about
everything, and
how breathing needs
love to sustain
it — a life spent
promised to uncertainty,
moonlight harvesting
my dreams between
fantasy and failure,
divine care letting
you slip away, a
moment lost each
time I remember
what might have
been true.

by Robert Darlington
from Poetry Feast

Rachel York in Guernica:

In his essay Sexual Objectification, Timo Jütten explains how sexual objectification teaches men and women to assume roles as superiors and subordinates, roles that disadvantage women and leave them vulnerable to gender-specific harms, such as sexual assault and rape. While I knew this already, it is a small relief to read it in print. Most women instinctively know that their bodies are a hair’s breadth away from violence. The artist Marina Abramović demonstrated this by laying seventy-two objects on a table—a tube of lipstick, a feather, a knife, and a gun—and invited gallery viewers to do whatever they wanted to her for six hours. At first the people were shy, presenting her with the flower, kissing her, draping her in cloth; then as the hours ticked off, one man snipped off her clothes with the scissors so that she was bare, and another sliced her with the knife. Someone else picked up the gun from the table, wrapped her hand around it, and pointed it at her neck. At the end she stood up to leave, bare, and bleeding, and the audience fled.

More here.