Thursday, November 30, 2017

How Olfaction Becomes You

AKA Osmetic Ontogenesis

In this jaw-dropping essay, the authors use the story of a sniffing salamander to explain the Uroboric cycle of perception. I know this isn’t a salamander, but it’s close enough, and cool as hell.

These sparkling crystals of olfactory knowledge are taken from the following essay:
Hosek R J & Freeman W J (2001). Osmetic Ontogenesis, or Olfaction Becomes You: The Neurodynamic, Intentional Self and Its Affinities with the Foucaultian/Butlerian Subject. Configurations 9: 509–541.

Our neurodynamic model of olfaction represents this situation. The organism does not directly perceive the impact of chemical odors from its surroundings on its receptor cells; rather, its unique perception consists in its prediction and hypothesis-testing into these surroundings. All that the animal perceives is the specific outcome of the tested hypothesis, and the activity of perception is formed in assimilation with the environment, which includes the organism’s state prior to and concurrent with the odorant impinging upon its receptors. These states consist in a variety of discourses, including the organism’s alertness, past experience with the odorant and all other aspects of its surroundings, and so forth. We can say, for example that a rabbit’s desire for a particular odorant is constructed within the discourses of the laboratory via the previously described odorant-training of newborn rabbits. Similarly, another organism’s desire for a certain cologne might be formed within advertising strategies that link particular scents to sexual pleasure and desirability. These desires correlate to brain states; thus, expressed in neurobiological terms, discourses form the shape of brain activity, its material substrate, and the emergence of intentionality in masses of living neurons. The odor’s meaning for and its very perception by the organism depends on these various factors, and this meaning and perception concomitantly shape the organism.

Friday, November 17, 2017

Why You Probably Don’t Know What Musk Really Smells Like

One of the big dogs of the perfume world is also a powerhouse of confusion. What the hell is musk? Do you know? Are you sure?

First, let’s disambiguate the basics. Musky and Musty are not the same thing. Musky is kind of like an animal smell – warm, sexy, intoxicating. Musty is related to moldy, it’s the smell of a dark hamper. Wet and dark, that’s musty. Thing is, body odor can be both musky AND musty, and this is one of the places where the two get intertwined. We can go on and on with this, because there’s this thing I call ‘river musk,’ which is absolutely intoxicating, to me at least, and is the by-product of stuff that lives in the river. Oceans have it too, but it tends to be mixed with dead fish. It’s a secondary metabolite, a kind of seaweed pheromone, which means it’s very similar to animal musk in that it’s a pheromone/chemosignal, but from plants. And yet, it’s related to wet things.

Parenthetically, body odor is actually not musty, for the most part. Usually it’s your dirty clothes that are musty, because they’ve been sitting in a pile on the floor, dark because they’re balled up, and damp because you just took them off, and your body is warm, and the air around you is less warm, so the water vapor in the air condenses on the clothes…either that or you were sweating just before you took them off. Either way, it’s your clothes that smell musty. That being said, we should also note that your armpits can get musty for the same reasons (it’s pretty dark in there). That isn’t what we call body odor, however.

Okay, so we got that out of the way. Sort of. Now for the anosmia part. Anosmia means you can’t smell something. Half the population is anosmic to something, and there are more people anosmic to musk than most other smells. Musk is a really big molecule, one of the biggest we can smell, and for some reason, bigger molecules tend to be invisible to us, but not in the way other anosmias work. It seems to go in-and-out.

Take the behemoth of the synthetic musks, Iso-E Super, which is so super that a perfume was made with only this as its only ingredient (and for non fragrance enthusiasts, that is totally unheard of). Lots of people can’t smell it. Saskia Wilson-Brown of the Institute for Art and Olfaction gave me a sample. I couldn’t smell it. Then I could. Then I couldn’t again. Anyway, because musk has this problem, or because we have this problem with musk, perfumers tend to use lots of musks in their formulas, to make sure that people will be able to smell at least one of the musks in there.

Next problem – the laundry detergent industry. Musks have this attribute where they do really well in fabrics and with detergents. They hold on for a long time (because they’re so big, as a molecule, among other reasons). They’re all over the world of laundry detergents; just about every laundry detergent smells like musk. But you probably don’t know that. You think that smell is the ‘smell of fresh laundry,’ not that of musk. When I tell people this, they don’t believe me, probably because they would rather not associate the smell of sweaty animal bodies with their freshly-laundered sheets. But alas, there it is: the smell of dirty is now the smell of clean.

What’s worse is that these sheets eventually end up in a dark hamper, and smell musty as well as musky. And so now you’re totally screwed. 

Friday, November 10, 2017

Ambiguity, Approximation and Probability

Logo (the programming language)

Probabilistic programming does in 50 lines of code what used to take thousands

I wanted to put some stuff up here about the state of computer programming, because the way we smell is akin to a special kind of computer program, and one which does not act like the kind we know.

I should start like this – I grew up on Logo, and then NES video games, therefore my experience with, and thinking upon, computer programming is ‘coded’ according to this top-down style. Someone writes the code, and the computer executes the code. There are no surprises (unless you have bugs to fix). You tell the turtle (that’s what they call the cursor in Logo) what to do and it does exactly that. Look at the picture above. That little triangle (the turtle) was told to go 100 spaces, rotate 90 degrees, then go 100 more spaces, etc, until a square is born.

King Koopa was told to jump every time you throw fireballs at him, or whatever he does. There is no fuzzy logic here. Everything is clear, concise, exact, predictable. (Again, that’s when the program runs as intended; surely this kind of programming is unpredictable when it goes wrong.)

Enter a new kind of programming. With the dual advent of big data and big processors to crunch it, we are seeing a different approach. The computing power is now so capable that it is asked to figure out its own program from the data given. This helps with a lot of the problems faced in computing today. With such variety in the data (this is ultimately what big data is about – not lots of quantity, but lots of different qualities) we can no longer write programs equipped to work with such variety. The program required for that ends up being as big as the dataset.

This is where we see the parallels to smells and olfaction. The amount of smells we could potentially be exposed to is infinite and multifarious. Vision has only a few categories. Things can look light or dark, a binary classification, or they can be categorized by their color on the spectrum, which is a discrete classification. They have a shape, a size, maybe a texture category. Odors, however, cannot be organized this way. There are too many and they are too different from eachother. Therefore, olfactory perception is distinct from our other senses. In order for us to create an artificial intelligence that can smell, we would have to come up with a different kind of programming.

Facial recognition provides a good visual analogy to the olfaction problem. What a face looks like isn’t really dependent on its color or its shape, but the combination of these features, the whole. And that makes a lot of initial parameters, in fact, infinite parameters. Face-rec uses these new types of programs, and they are almost the opposite, in every way, of what programming has been. I’ll let this guy describe them:

“When you think about probabilistic programs, you think very intuitively when you're modeling. You don't think mathematically. It's a very different style of modeling.” … “The code can be generic if the learning machinery is powerful enough to learn different strategies for different tasks.”
- Tejas Kulkarni, an MIT graduate student in brain and cognitive sciences,

In the same way that we are not born already knowing every smell we will ever encounter, these programs must ‘learn on the fly.’ This is an advance in computing, but also it foreshadows a very different world, where information is not distinct, discrete, exact, etc. It is instead more like that thing you smell but you don’t know what it is, but you swear you know yet you don’t know…you know what I’m talking about? Doesn’t sound like the kind of output your computer would produce.


[lots of good explaining in this article, so I just copied most of it]

A Grand Unified Theory of Artificial Intelligence

Embracing uncertainty

In probabilistic AI, by contrast, a computer is fed lots of examples of something — like pictures of birds — and is left to infer, on its own, what those examples have in common. This approach works fairly well with concrete concepts like “bird,” but it has trouble with more abstract concepts — for example, flight, a capacity shared by birds, helicopters, kites and superheroes. You could show a probabilistic system lots of pictures of things in flight, but even if it figured out what they all had in common, it would be very likely to misidentify clouds, or the sun, or the antennas on top of buildings as instances of flight. And even flight is a concrete concept compared to, say, “grammar,” or “motherhood.”

As a research tool, Goodman has developed a computer programming language called Church — after the great American logician Alonzo Church — that, like the early AI languages, includes rules of inference. But those rules are probabilistic. Told that the cassowary is a bird, a program written in Church might conclude that cassowaries can probably fly. But if the program was then told that cassowaries can weigh almost 200 pounds, it might revise its initial probability estimate, concluding that, actually, cassowaries probably can’t fly.

“With probabilistic reasoning, you get all that structure for free,” Goodman says. A Church program that has never encountered a flightless bird might, initially, set the probability that any bird can fly at 99.99 percent. But as it learns more about cassowaries — and penguins, and caged and broken-winged robins — it revises its probabilities accordingly. Ultimately, the probabilities represent all the conceptual distinctions that early AI researchers would have had to code by hand. But the system learns those distinctions itself, over time — much the way humans learn new concepts and revise old ones.

Thursday, November 9, 2017

Getting Schooled on Deep Learning

This image was produced by a neural network called the Deep Dream Generator

Despite their great usefulness, deep learning neural nets have been unexplained in terms of how they do what they do.

New discoveries are being made, however, that shed light inside the black box of these networks:

Oct 2017, WIRED

In one case, the researchers used small networks that could be trained to label input data with a 1 or 0 (think “dog” or “no dog”) and gave their 282 neural connections random initial strengths. They then tracked what happened as the networks engaged in deep learning with 3,000 sample input data sets.

This is an example of an experiment using a network of only couple hundred nodes. All the nodes have been tagged and watched to see how they evolve over iterations of the network. In other words the network is given a task, let's say to recognize face in a picture, and it attempts that task over and over.

The special thing about neural networks is that they learn how to do this task better with every iteration. When our brains do this we call it trial-and-error. We learn how to do things by trying over and over again, and hopefully we get better. These networks try over and over and eventually get better at what they're supposed to be doing.

What's happening while they try has been unknown, or we could say that it still is unknown. But experiments like this are helping us to learn more.

This is a good moment to recall that olfaction, or to be more specific – the olfactory bulb – is a biological neural network. The olfactory bulb is the brain of the nose; it is the nexus at which molecules in the air are translated into electrical signals that the brain can recognize as a smell.

The brain is a very large neural network. The olfactory bulb, on the other hand, is a very good model of this neural network phenomenon. It can be teased apart and separated from the rest of the brain very easily because it functions as its own brain.

Before we had a brain – the human part, the cortex, the one that helps to read this text – we had only a nose and the nose-brain. It can be said, in the spirit of metaphor of course, that the olfactory system was the first cortex, it was built on top of the limbic system, and it still connects to the limbic system in its own separate way.

Smell works differently than all the other senses; it has a direct line to the limbic system, which is the command center of the brain. It has a direct line to the White House, in other words...

So as we discover what neural networks are really doing in their hyperconnected webs, let us remember that the olfactory system did it first.

Post Script

Friday, November 3, 2017

Urban Scentsations

Odor investigators engaging in a smell hunt. source

Let’s take a minute to recognize this exceptional olfactory artist, Kate McLean. She works with human perception and the urban smellscape. She basically turns the city into her own little laboratory, running perception experiments on the people there, and coming back with sensory analyses that you just can’t get in an actual lab setting. She does smellwalks, smell sketches, and all kind of other activities to both help people explore and appreciate the most overlooked aspect of their urban environment. As a result of her work, we get these Smellmaps, something not taken up by Google yet…yet. Please check out her site here, and get upset that she already came to your neighborhood, or excited that she didn’t yet!

Research, analysis & design of Sensory Maps by Kate McLean