Showing posts with label art vs science. Show all posts
Showing posts with label art vs science. Show all posts

Friday, April 8, 2022

This Side of Science


Science, it's full of surprises. But I am most excited when the scientists themselves are the ones surprised. The two most exciting words I can read in the news are "scientific accident", and you'll see them below. 

First up, a long-time mystery of the universe, swept under the rug by Einstein himself, some might even call it Einstein Fudge. You know it as the cosmological constant, and it says that the universe is constantly expanding. We don't understand why the cosmological constant is there or what it's really doing, but we can't make sense of the universe without it. Until now. 


New study proposes expansion of the universe directly impacts black hole growth
Nov 2021, phys.org

Black holes grow along with the expansion of the universe. Black holes gain mass from the expansion of the universe itself. Masses of black holes could grow in lockstep with the universe, a phenomenon that Croker and his team call cosmological coupling.

The most well-known example of cosmologically-coupled material is light itself, which loses energy as the universe grows. "We thought to consider the opposite effect," said research co-author and UH Mānoa Physics and Astronomy Professor Duncan Farrah.

"It was a such a simple idea, I was surprised it worked so well."

via University of Hawaii at Manoa, University of Chicago, and University of Michigan at Ann Arbor: Kevin S. Croker et al, Cosmologically Coupled Compact Objects: A Single-parameter Model for LIGO–Virgo Mass and Redshift Distributions, The Astrophysical Journal Letters (2021). DOI: 10.3847/2041-8213/ac2fad


Granddaughters and great-granddaughters of men who start to smoke before puberty, have more body fat than expected
Jan 2022, phys.org

I'm just saying, how the do you even figure this out. Who is sitting there right now putting this hypothesis together in their head.

via University of Bristol: Human transgenerational observations of regular smoking before puberty on fat mass in grandchildren and great-grandchildren, Scientific Reports (2022). DOI: 10.1038/s41598-021-04504-0


What we knew about water was right after all
Feb 2022, phys.org

Like the discovery of lead in the air - when something is everywhere, it's hard to notice it at all.  In this case, it seems the fact that this experiment was never done in a desert environment is all that was keeping us from the answer. 

Frankenpaste:

Probing chemical transformations at the air-water interface is challenging due to the lack of surface-specific techniques or computational models, but now they found water spontaneously transforms into 30–110 micromolar hydrogen peroxide (H2O2) as microdroplets. The textbook understanding of water is thus challenged by how the mild temperature and pressure conditions, together with the absence of catalysts, co-solvents and significant applied energy, could break covalent O–H bonds; it was previously thought to be from ultrahigh electric fields. 

It was ambient ozone -- Although ozone minimally dissolves in water, the enhanced surface area of microdroplets allows more ozone to be dissolved and quickly react to form H2O2.

"There had to be something related to the geography of the place, an environmental difference between our location in Saudi Arabia and California," Gallo Jr says.

via King Abdullah University of Science and Technology: Nayara H. Musskopf et al, The Air–Water Interface of Water Microdroplets Formed by Ultrasonication or Condensation Does Not Produce H2O2, The Journal of Physical Chemistry Letters (2021). DOI: 10.1021/acs.jpclett.1c02953


Life may actually flash before your eyes on death
Feb 2022, BBC News

First-ever recording of a dying brain discovered by accident. 

"This was actually totally by chance, we did not plan to do this experiment or record these signals."

This is also one of the reasons why it is important to care for every human equally, regardless of what happened to them. You're born without an immune system? We're keeping you alive. Paraplegic? We're giving you wifi for your body

You have epilepsy? We're going to slap some electrodes to your head and monitor your brainwaves for a really, really long time, and figure out how to help you. Unless you have a heart attack in the headset, in which case we'll watch what happens, and use your accident to further the advancement of science. 

via Department of Neurosurgery, Henan Provincial People’s Hospital, Division of Neurosurgery, Vancouver General Hospital: Vicente Raul et al. Enhanced Interplay of Neuronal Coherence and Coupling in the Dying Human Brain. Frontiers in Aging Neuroscience 14 2022. DOI: 10.3389/fnagi.2022.813531.


Researchers report game-changing technology to remove 99% of carbon dioxide from air
Feb 2022, phys.org

Just some very clever thinking. 

Hydroxide exchange membrane (HEM) fuel cells, an economical and environmentally friendly alternative to traditional acid-based fuel cells used today, have a shortcoming that has kept them off the road—they are extremely sensitive to carbon dioxide in the air. Essentially, the carbon dioxide makes it hard for a HEM fuel cell to breathe. This defect quickly reduces the fuel cell's performance and efficiency by up to 20%, rendering the fuel cell no better than a gasoline engine.

"Once we dug into the mechanism, we realized the fuel cells were capturing just about every bit of carbon dioxide that came into them, and they were really good at separating it to the other side," said Brian Setzler, assistant professor for research in chemical and biomolecular engineering and paper co-author.

While this isn't good for the fuel cell, the team knew if they could leverage this built-in "self-purging" process in a separate device upstream from the fuel cell stack, they could turn it into a carbon dioxide separator.

They found a way to embed the power source for the electrochemical technology inside the separation membrane. The approach involved internally short-circuiting the device.

via University of Delaware: Lin Shi et al, A shorted membrane electrochemical cell powered by hydrogen to remove CO2 from the air feed of hydroxide exchange membrane fuel cells, Nature Energy (2022). DOI: 10.1038/s41560-021-00969-5


Unrelated Post Script:
Parasite that replaces a fish's tongue caught at Texas state park
Oct 2021, phys.org

The tongue-eating louse makes its way into the fish's mouth through its gills, where it consumes the mucus that forms on the inside of the fish's mouth, and completely replaces the organ in another creature, although it doesn't harm its host.

Also, it has eyes; the parasitic zombie tongue has its own eyes.

Monday, February 17, 2020

Deep Creep

Visualizing and Understanding Convolutional Networks -- This goes back to 2013 already, but I recently came across these images, and they are so mesmerizing I had to archive them.

The images pasted here are from a paper about convolutional neural networks. The researchers are able to train a network with tons of images, and then ask that network to classify new images it hasn't seen yet. Getting the detection error rate down to zero is the goal. This one does a good job.

But what makes this report special, is that we get to see how the system spits back what the different "neurons" see. The network develops layers or clusters that recognize different things; some are good at low-level features like lines and edges, and some are good at high level things like "bicycles" or "origami." Together they learn how to see.


^This is the first layer, it sees angles and colors.


^This is the second. This one's getting more complicated patterns. Notice the similarities, but also the differences. Of the 3x3 sets, which one is not like the other? The network saw all of those as similar. The network says that those images sit close together in image-space (the total space of all possible images, or at least all the images it was trained on).

It's a game to try and figure out the common denominator. We don't really know the common denominator, and we can't know, because we're just not computers. This is why they call it "deep." In the middle of this network, deep in there, is a layer with information that is just too complex for us. Collapse 33,000 dimensions into 3, and now try and communicate features of the 33,000 using only those 3 points of information. Can't do it. That's why it's mysterious.


^Now the third layer. Ok, so I see the people, I see the barcode/text motif, the honeycomb/diamond clusters, even the lady bug and the tomato -- it's a stretch, but I get it. But some of these are just nuts. White lower right corners? It sees white lower right corners?


^Fourth layer. No idea what this thing is talking about.


^Fifth layer. Here we go, people, dogs, flowers, now it's all making sense. And that was the point of creating this network -- you give it any picture of a flower, no matter how weird of a picture, barely looks like a flower, and this network will recognize it and classify it as a flower. (Mostly; it's not perfect.)

But there comes a point in the middle there, we have no idea what this thing is thinking about. And we never will. Just like other people, and how we can never really know what someone else is thinking. Which is to say that the computers are now a lot more like other people than they ever used to be. They have become mysterious.

Notes:
Zeiler and Fergus, Visualizing and Understanding Convolutional Networks, 2013 [ZFNet].
https://arxiv.org/abs/1311.2901

Saturday, November 3, 2018

On Science Communication




Science isn't easy. It can be repetitive and tedious, but it's not easy. Writing about it is even harder.

We're skimming through a grocery list of the worst scientific disciplines to research and write about, as compiled by the writers of Ars Technica, a science-slanted popular news outlet.

Note that this list is more about what science writers don't like --reading-- about, not writing. And most of the problems here come from the discrepancy between how passionate the writers feel about their subject matter, and its impenetrability.

Space exploration
Paleontology
Batteries
Astronomy
Theoretical physics
Archeology
Quantum optics e.g. time crystals

The reasons for this lamented impenetrability are succinctly summarized by an actual scientist, in the comments section:
Professional scientist here. I'm currently writing a review paper, and I'm on a bit of a deadline crunch, so naturally I'm procrastinating here on Ars. And regarding your accusations of the dryness and joy robbing nature of the primary literature, I have just one thing to say to you...
You're kinda right. I mean, sure, there's a reason that it's written this way, but yeah, I get it. I'm plowing through many papers in my field and I understand the details well, and I definitely get excited at many passages that would put most people to sleep. But many simply put me to sleep.
I think the reason for this is three-fold: 1) precision of language is important; 2) making precise language varied and interesting is much more work than allowing it to remain dry, so many authors don't do it; 3) there are some details that specialists need to communicate to each other that aren't exciting no matter how hard you try. ¯\_(?)_/¯
-jjemerson

This is a great explanation, and supports the need for this very specific type of communicator in our society, the science writer. Alan Alda thought it was so important that he created an entire school to attend this need.


Notes:

Here are the subjects our reporters enjoy covering the least
Sep 2018, Ars Technica

Alan Alda Center for Communicating Science at Stony Brook University





Friday, October 12, 2018

Entropy Goggles


A computer has been taught to distinguish between different arrangements of the elements and principles of design across artists and over time. But don't worry, it's nowhere near human.

Much of what makes this possible is the amount of artwork that has been digitized, ie, machine readable, ie, machine consumable. We can now feed our computers the same diet of cultural nutrients that we as photosensitive creatures enjoy, giving it a fair chance.

The computer can't see the things we see in a work of art. It can't see the richness and it can't assign meaning anywhere near the way we can. But it can do some things even better than us, especially if we give it a concise metric with which is can "familiarize" itself with the artwork.

The primary metric used here was one of complexity-entropy, where each pixel in an image was measured in terms of the complexity of its spatial patterns. A fully white square, or a square filled with all manner of colors and lines. Each pixel value is then added to give the picture an overall score.

The score of the different artworks in the database shifted in tandem with major accepted periods of artistic periods, making it look like this program has figured something out, can recognize something about our artifacts of cultural assimilation.

Machine learning enables physics-inspired metrics for analyzing art
Aug 2018, phys.org