Plume Tracking and Odor Mapping Algorithms

A deep reinforcement learning model that allows AI agents to track odor plumes

Feb 2023, phys.org

https://techxplore.com/news/2023-02-deep-ai-agents-track-odor.html

Insects track odor plumes to find mates. (And there was a similar study done recently here.)

"Instead of running a traditional laboratory wind-tunnel experiment, we used a complementary 'in-silico' approach using artificial neural networks," Singh explained. "This helped us develop an integrative understanding of plume tracking across multiple levels, including emergent behavior, neural representation and neural dynamics."

To train their plume-tracking agents using DRL, the researchers first simulated an odor emanating from a source located within a windy arena with a total area of approximately 120 m2. When their agents identified where the source of the odor was located, they received a reward. In contrast, if they lost track of the odor plume and left the arena, they were "punished."

"The behavior that emerges in our trained artificial agents bears a striking resemblance to the behavior modules biologists have previously observed in flying insects performing plume tracking," Singh said.

via University of Washington and University of Nevada: Satpreet H. Singh et al, Emergent behaviour and neural dynamics in artificial agents tracking odour plumes, Nature Machine Intelligence (2023). DOI: 10.1038/s42256-022-00599-w

https://dx.doi.org/10.1038/s42256-022-00599-w

Image credit: AI Art - Total Hallucination - 2023

Simulated Oversight

'Virtual nose' may reduce simulator sickness in video games

Purdue University News, Mar 2015  

https://www.purdue.edu/newsroom/releases/2015/Q1/virtual-nose-may-reduce-simulator-sickness-in-video-games.html

Yes, you can relieve motion sickness in virtual reality by coding a nose overlay into the frame as a visual guide.

I sure didn't realize it until I read this, but you're looking at your nose all day; you're looking at it right now. Well, maybe you're not looking at it, but it's there. Maybe now you're looking at it, since we're talking about it. Anyway, it's there all day. If your eyes are open, your nose is in your field of vision. And when it's not, you're disoriented.

I'm pretty sure this is not what they mean when they say "right under your nose" or "right in front of your face," but it sure works in this case. Throw that onto the heaping pile of other things we don't notice about our nose or what it does for us. And add that to the other pile of things that we could improve if we looked to the nose and olfaction in general as a source of biomimetic supremacy.

Notes:

via Purdue University: Whittinghill, D.M. et al. Nasum virtualis: A simple technique for reducing simulator sickness. In Proceedings of the Games Developers Conference (GDC), San Francisco, CA, USA, 2–6 March 2015; p. 74.

via University of Wuerzburg: A Virtual Nose as a Rest-Frame - The Impact on Simulator Sickness and Game Experience. Carolin Wienrich et al, 10th International Conference on Virtual Worlds and Games for Serious Applications, Sep 2018. DOI:10.1109/VS-Games.2018.8493408

https://doi.org/10.1109/VS-Games.2018.8493408

via the BioComputing Lab at Korea University of Technology and Education: A Study on Visually Induced VR Reduction Method for Virtual Reality Sickness. Ju-hye Won and Yoon Sang Kim. Appl. Sci. 2021, 11(14), 6339; https://doi.org/10.3390/app11146339.

Flipping the Switch

The 'surprisingly simple' arithmetic of smell

Jan 2022, phys.org

https://medicalxpress.com/news/2022-01-surprisingly-simple-arithmetic.html

The age of artificial olfaction is upon us.

This is now the second report in the last few months that presents a computational model for the olfactory bulb, which is the biological supercomputer on your face that crushes gigtons of databytes per attosecond (slight exaggeration).

The last paper came from a physicist working on information theory (Tavoni et al at Penn State). Another paper the month prior, which came from none other than the lab that discovered olfactory receptors, found, again, a computational model, discovered via machine learning, that compresses the n-dimensionality of odorant sensory data. 

But again, this new paper comes primarily from a department of electrical and systems engineering, in collaboration with the biomedical engineering department. I don't know everything that's going on, I only read the papers on the weekends, but that's a lot of papers about computing in olfaction, and from people who do not study olfaction exclusively.

And this is at the level of the bulb. We're not talking about the DREAM project, where big-data's worth of words and molecules are processed by GPT-3 to predict the names of smells. This is about looking at the hardware. How in the world does that bulb, which compresses thousands of receptors, themselves receiving information from un-countable stimuli, into dozens of signals that go on to control the entire enormity of a mammalian body via its limbic system. The bulb is the choke point for this system, and it's using magic that we are only now beginning to understand to the point of copying it. 

I doubt this is the earliest example, but as far back as 1991, scientists were talking about olfaction as a model system for computational neuroscience. These were neuroscientists and psychologists writing about this. But they could see the significance -- it's literally wired like the deep learning neural networks you hear about in the news (you know, powering the AI in your toaster, your tissue box and your alarm clock). 

It really looks like we're getting the hang of this. They started with a simple question -- how come things smell the same to us, even in different contexts or environments? Like how a plaid shirt looks okay in your sunlit bedroom, but later looks like a shit sandwich in the fluorescent lights of your office (or remember the black-and-gold dress? maybe you're trying to forget). 

If smells come from evaporating molecules, which are literally volatile, changing all the time based on environmental conditions, how come they always smell the same to us? Maybe olfaction would be a good model to investigate. 

So they did, by pairing locusts with a training smell, under all kinds of different conditions, hungry, full, hot, cold, humid, dry. Every time, the locust recognized the training smell (with the locust equivalent of a salivating dog). Yet, "The neural responses were highly variable," one of the researchers said. Same molecule, same response, but completely different receptor patterns, every time. It just doesn't make sense.

Deep learning to the rescue (obviously). The algorithm found that it's the interaction of activating and inhibiting neurons; I'll copy the copy directly:

Finding the features you want is similar to the information conveyed by the ON neurons. Absence of deal breakers is similar to silencing of the OFF neurons. As long as enough ON neurons that are typically activated by an odorant have fired—and most OFF neurons have not—it would be a safe bet to predict that the locust will open its palps in anticipation of a grassy treat.

via the Department of Electrical and Systems Engineering and the Department of Biomedical Engineering, Washington University in St. Louis: Srinath Nizampatnam et al, Invariant odor recognition with ON–OFF neural ensembles, Proceedings of the National Academy of Sciences (2022). DOI: 10.1073/pnas.2023340118

http://dx.doi.org/10.1073/pnas.2023340118

And further reading:

via University of Pennsylvania: Gaia Tavoni et al, Cortical feedback and gating in odor discrimination and generalization, PLOS Computational Biology (2021). DOI: 10.1371/journal.pcbi.1009479

http://dx.doi.org/10.1371/journal.pcbi.1009479

via Massachusetts Institute of Technology's McGovern Institute for Brain Research: Peter Y. Wang et al, Evolving the olfactory system with machine learning, Neuron (2021). DOI: 10.1016/j.neuron.2021.09.010.

http://dx.doi.org/10.1016/j.neuron.2021.09.010

via MIT: Davis J L & Eichenbaum H, eds. (1991). Olfaction: A Model System for Computational Neuroscience. Boston: Bradford Books/MIT Press.

https://mitpress.mit.edu/books/olfaction

Deep Nose, 2022

http://www.limbicsignal.com/2022/02/deep-nose.html

Signal to Noise for the Win, 2021

http://www.limbicsignal.com/2021/11/signal-to-noise-for-win.html

Olfatory Overload, 2021

http://www.limbicsignal.com/2021/05/olfactory-overload.html

Image credit: Inhibitory Synapse - TAO Changlu, LIU Yuntao, and BI Guoqiang; Image design: WANG Guoyan, MA Yanbing - 2021

Deep Nose

Artificial networks learn to smell like the brain

Oct 2021, phys.org

https://medicalxpress.com/news/2021-10-artificial-networks-brain.html

We now have an artificial neural network that works like the nose. It's not an electronic nose; that's not the big deal part. 

What's important here is that, first of all, this is from the lab that brought us olfactory receptors. Next, these scientists didn't even model the network on evolution. They made an algorithm to solve an odor categorization task, and let the network run with it. Some might call that artificial evolution, but scientists will call it machine learning. After several iterations, the network found an optimized structure for solving this task -- the network ended up looking just like our olfactory system. Go figure.

The "brains" of a neural network lies in its ability to reduce the dimensionality of the information thereby optimizing computation. This is done using compression layers that learn to accept information from some neurons and not from others. After enough iterations, a pattern emerges between the layers of neurons.

Talking about this pattern, and the number of neurons connected to by each neuron on the compression layer:

"It could have been one, it could have been 50. It could have been anywhere in between," Yang says. "Biology finds six, and our network finds about six as well."

-Guangyu Robert Yang, associate investigator at MIT's McGovern Institute for Brain Research, and postdoc at Columbia University

The first part of our nose where the hundreds of olfactory receptors collapse into far less neural nodes, one layer up the network, is very similar to the type of artificial neural network used in this study. It is, apparently now supported by these findings, a very effective way to condense a multi-dimensional information-space of chemical signals. Would it work for other information-spaces? What other Big Data can this deep nose model compute?

The side story: Evolution found this organization through random mutation and natural selection over eons; the artificial network found it through standard machine learning algorithms, in under one minute. 

via  Massachusetts Institute of Technology: Peter Y. Wang et al, Evolving the olfactory system with machine learning, Neuron (2021). DOI: 10.1016/j.neuron.2021.09.010

http://dx.doi.org/10.1016/j.neuron.2021.09.010

Image credit: Google's quantum computer, totally unrelated, just for looks

Post Script:

Neural network reveals new insights into how the brain functions

Dec 2021, phys.org

https://medicalxpress.com/news/2021-12-neural-network-reveals-insights-brain.html

"The neural network model approach we have developed in this work presents an 'instruction manual' for other researchers to use to study other areas of the brain or other organs"

-co-author Dr. James Martin, co-author and professor of molecular physiology and biophysics at Baylor College

Their model is called Spatial Transcriptomics cell-types Assignment using Neural Networks (STANN).

via Baylor College of Medicine: Francisco Jose Grisanti Canozo et al, Cell-type modeling in spatial transcriptomics data elucidates spatially variable colocalization and communication between cell-types in mouse brain, Cell Systems (2021). DOI: 10.1016/j.cels.2021.09.004

http://dx.doi.org/10.1016/j.cels.2021.09.004

I'm not certain about this, but I think the reason they chose the olfactory system is because they were looking at the interaction of transcription genes and brain cell types, and the olfactory receptor brain cells are the only one's that each get their own gene. So that would make the olfactory bulb an ideal nexus for investigation of this kind. See studies below for further reference:

Marei H.E.S. et. al. Gene expression profile of adult human olfactory bulb and embryonic neural stem cell suggests distinct signaling pathways and epigenetic control. PLoS One. 2012; 7: e33542. https://doi.org/10.1371/journal.pone.0033542

Nagayama S. et. al.  Neuronal organization of olfactory bulb circuits. Front. Neural Circuits. 2014; 8: 98. https://pubmed.ncbi.nlm.nih.gov/25232305/

Olfaction In Silico

Artificial networks learn to smell like the brain

Nov 2021, phys.org

https://medicalxpress.com/news/2021-10-artificial-networks-brain.html

No kidding, the sensory apparatus that resembles a deep learning neural network can be simulated with a deep learning neural network -- "Artificial networks trained to classify odor identity recapitulate the connectivity inherent in the olfactory system."

-Neuron

The part of our brain that smells is also the most primitive. Before brains were a thing, bacteria performed chemosensory calculations on the primordial soup. As the soup became more complex, so did sensory equipment. Chemosensitive receptors on the surface of a bacterium became antennae, and then became noses, and those noses became seeing, hearing, even speaking brains. But the first version is the one used for smelling. So it shouldn't be a surprise that the first place we see a direct link between the mammalian brain and our artificial instantiation is via olfaction. Nonetheless, the scientists were "surprised to see it replicate biology's strategy so faithfully."

-medicalexpress

"By showing that we can match the architecture very precisely, I think that gives more confidence that these neural networks can continue to be useful tools for modeling the brain," says Robert Yang, assistant professor in MIT's departments of Brain and Cognitive Sciences and Electrical Engineering and Computer Science [and who collaborated on this project with Columbia neuroscientists Richard Axel and Larry Abbott, btw].

-medicalexpress

They use an antennae model, and the indispensable fruit fly, but it's all close enough to humans, I mean that's why we use the fruit fly in the first place. They started with some artificial neurons, of the same amount found in a fruit fly. They programmed the neurons to identify odors, and to assign valence (pleasant or unpleasant) to odors. They didn't give the neurons any structure, no information about how to talk to each other, no blueprint on how to process information. Just pre-programmed neurons, thrown into a simulated universe of chemosensation.

In minutes, and in silico, a network emerged to look just like the nose-brain of a fruit fly. All on its own, "an initially homogeneous population of neurons segregated into two populations with distinct input and output connections, resembling learned and innate pathways." 

-Neuron

In other words, it learned how to smell. It took evolution some billions of years to get the fruit fly olfactory system just right. The artificial network did it in minutes. Extrapolations from the study abstract: "This implies that convergent evolution reflects an underlying logic rather than shared developmental principles."

-Neuron

-Image credit: Chris Thomasson via Fractal Forums - Expanded - 2017

*Their structure used expansion and compression layers, similar to the pyramidal-structure of the nose-brain, and the number of inter-connections was also the exact same number as in the fruit fly, and it worked with both feedforward and recurrent network models, and the networks are plastic, meaning they can learn new odor associations over time.  

via Massachusetts Institute of Technology's McGovern Institute for Brain Research: Peter Y. Wang et al, Evolving the olfactory system with machine learning, Neuron (2021). DOI: 10.1016/j.neuron.2021.09.010

http://dx.doi.org/10.1016/j.neuron.2021.09.010

Post Script:

The book Hidden Scents talks about how olfaction is an ideal model for understanding an artificial brain, for an artificial human. 

Although artificial neural networks resemble natural neural activity patterns, like those used by the visual cortex for object recognition, we still don't understand how the visual cortex, or most mammalian neural circuits, are inter-connected. This time, we see how they connect, a connectome of the olfactory cortex.

Post Post Script:

And this study should be kept alongside this other one, where a physics-based computational neuroscientist came up with a pretty simple way to mimic the olfactory cortex, via University of Pennsylvania: Gaia Tavoni et al, Cortical feedback and gating in odor discrimination and generalization, PLOS Computational Biology (2021). DOI: 10.1371/journal.pcbi.1009479

http://dx.doi.org/10.1371/journal.pcbi.1009479

Neuromorphic Odor Translator Helps Robots Express Their Feelings

Neural network trained to properly name organic molecules

Aug 2021, phys.org

https://techxplore.com/news/2021-07-neural-network-properly-molecules.html

Do you ever have a problem naming that smell? It's not just you. Science has this problem too, but maybe not for long. 

Smells are volatile organic compounds that have evaporated and entered your nose. Although they almost always occur in combination with others and not in isolation, us humans want to reduce smells to their individual components, and then name them. After all, in order to think about something, you have to know it's name. (Is that true?) 

The problem is that organic molecules are big, with lots of chemicals joined together in lots of ways, so coming up with a naming convention for all these permutations is hard. IUPAC, the International Union of Pure and Applied Chemistry, sets the convention for naming molecules. And boy is it complicated.

Take sugar, a common molecule known to us by its simple name "sucrose;" in IUPAC, it's called (2R,3R,4S,5S,6R)-2-[(2S,3S,4S, 5R)-3,4-dihydroxy-2,5-bis(hydroxymethyl)oxolan-2-yl]oxy-6-(hydroxymethyl)oxane-3,4,5-triol.

Since we do live in the computer age, folks want to automate this naming process for when they discover new molecules. But as you can imagine by looking at the IUPAC name for sucrose, the algorithm at the core of that naming convention is really hard to write. So they decided to use a neural network instead.*

*I'm casually calling a neural network "neuromorphic," but in the past few years, real neuromorphic computers have forced a distinction here that I'm ignoring here for the sake of a more clickable title. 

This new artificially intelligent chemical translator is not a magical structure-to-name translator that can just look at a chemical and give it a name; that's still out of reach. It does, however, translate between IUPAC and another naming convention called SMILES.

Trained on PubChem's 100 million molecules, this translator ultimately shows how the utility of the new approach of using neural networks to help us write algorithms from the bottom up instead of the top down, which really is a revolution in computing. 

And if you think it would be cool to have robots that can smell, or to ensure that future humans maintain their sense of smell as they evolve into hyperdimensional algorithms, then making odors machine-readable is how you do that.

via Skolkovo Institute of Science and Technology, Lomonosov Moscow State University and start-up Syntelly: Lev Krasnov et al, Transformer-based artificial neural networks for the conversion between chemical notations, Scientific Reports (2021). DOI: 10.1038/s41598-021-94082-y

Artificial Olfactory Perception and the Olfactome

Chemical informatics, machine learning and the indispensable fruit-fly, Drosophila melanogaster have been used by researchers at University of California Riverside to predict odor perception. 

Olfactory prediction is kind of a holy grail of sensory perception. Sounds sus. Let's get into the data.

image credit: Diatom, by Dr. Jan Michels for Nikon Small World 2020

Using artificial intelligence to smell the roses

Aug 2020, phys.org

https://phys.org/news/2020-07-artificial-intelligence-roses.html

First sentence they're referencing Asifa Majid. That's a great start. Her work shows us that culture, language and experience influence individual odor perception. Nonetheless, the search for the human odor code continues.

After reducing a larger dataset of 84 olfactory receptors and 54 allelic variants (138 total), they took 34 receptors, each of which is controlled by a single gene, and trained machines to predict their descriptors. The descriptors, or "the words we would use to describe the smell," came from the Vosshall Keller Rockefeller University 2016 lexicon. They've got about 170 odorants, working on 34 receptors. 

Remember that each odor receptor gene can be activated by a number of chemicals, sometimes by only one, but usually by more than one. This is what makes things complicated. Olfaction is a combinatorial affair that breaks down at the granular level.

And they made a model for each receptor, 34 different models, and fed those models the odorants. They found that you could predict chemical properties of the molecules that match each receptor tested. So now, we can use the 450,000 library of chemicals, run them through each of the 36 artificial receptors, and predict what those receptors would perceive.

Figure 5A: Few Key ORs or Chemical Features Sensibly Cluster the Perceptual Descriptors

(A) Dendrogram representation of the Euclidean distances among perceptual descriptors based on overlap of perceptual response data (% Usage) from chemicals in the ATLAS study.

(B) Dendrogram from the top five ORs picked per perceptual descriptor.

(C) Dendrogram created from five randomly chosen ORs per perceptual descriptor.

(D) Dendrogram from the five best overall predictors including OR and chemical features per perceptual descriptor. Clustering is hierarchical and based on Euclidean distance (A) or the Jaccard distance (B–D). Cluster number (colored branches) inferred from gap statistic across bootstrap samples. [find the pdf for fine-resolution]

-source

I think, and I could be wrong, but it seems the big deal here is that they made a model for each receptor, instead of just making one model for all receptors. Whereas others have created an n-dimensional predictive space to collapse the behemoth of the chemosphere into a single equation, this team just reverse-engineered the receptors themselves.

They haven't found the odor code, but they did write 34 of them. We have hundreds of olfactory receptors. That's not everything, but we are definitely getting there.

What it CAN do? It can help us discover new chemicals, and also to discover substitutes for other chemicals that are expensive, rare, or ethically-troublesome (fear-pheromones from tortured cats for example).

What it CAN'T do? It can't predict how an odor will smell to you, as an individual. It can approximate, however, and pretty good. They mention only getting 20% of the human olfactome, or human olfactory receptor repertoire.

via UC Riverside: Joel Kowalewski et al. Predicting Human Olfactory Perception from Activities of Odorant Receptors, iScience (2020). DOI: 10.1016/j.isci.2020.101361

http://dx.doi.org/10.1016/j.isci.2020.101361

Post Script:

They mention something called the ATLAS dataset, but I don't know what that is, other than a proprietary data analysis software. Maybe it's their own dataset through ATLAS?

And for fun, I'll report that they do mention "substantive portion of odor identity arises early in the processing stream" which is a good way of describing the the two-layer perception process of olfaction.

The second layer, and this is the one that Asifa Majid tells us is influenced by culture, experience, and language: "It is likely that the remaining portion depends on experience-dependent modulation, supporting a downstream model with reliance on distributed neuronal networks for human perceptual coding."

Further: "Unlike the retinotopic and tonotopic patterning observed in the visual and auditory cortices, representing spatiotemporal properties of visual and auditory stimuli as they are processed at sensory neurons, piriform activity appears randomly distributed, without a clear mapping of physicochemical features (Stettler and Axel, 2009)."

Interesting: "In our analyses, the OR specialized for musk was not a top candidate for

musk predictions but contributed strongly to predictions of 'sweaty.'"

Perhaps because the model isn't "smelling" it among other calculated fragrant mixtures such as perfumes, but rather "in the wild?" 

Post Post Script:

Can't talk about the odor code without mentioning code smell, a term for when something is wrong with your code, but we're not sure what it is. 

https://en.wikipedia.org/wiki/Code_smell

Also, going deep on the topic here:

The Dream of Olfaction Prediction

http://www.limbicsignal.com/2020/01/the-dream-of-olfaction-prediction.html

The Future of Perception

How neural networks work—and why they’ve become a big business

Dec 2019, Ars Technica

Graphics cards, backpropagation, and big data – the ingredients for the deep learning revolution.

Our sense of smell is a great example of a deep-learning-style brain module, or to use a fashionable term, brain ensemble. The processing center of olfaction is the piriform cortex, the structure of which looks pretty similar to the pyramid diagrams of neural networks (and hence the origin of its name).

Furthermore, our olfactory system works a lot like a black box – we don't know what it's doing in there. The receptors do not code molecules one-to-one and they don't seem to correlate to perception in a meaningful or predictable way.

Our autobiographical memory, which is controlled and enriched by our olfactory perceptions, is a pretty big dataset. It holds all your personal memories, but including the physiological data as well, such as heart rate, hormone patterns, even sensorimotor data.

Now I'm not sure where the graphics cards come into the picture, something about parallel processing I guess. Olfaction uses around 400 receptors and vision uses 4 (three cones and one rod). I'm not a computer scientist, but GPU is like the opposite of CPU (in this context of neural nets) just as parallel processing would be opposite to serial processing. I’ll have to let someone else articulate that analogy.

Anyway, graphics cards, backpropagating neural networks, and big data are building the "deep" revolution, and I am waiting for the day that olfaction takes its role as the model sense for helping us understand and interact with our omnipotent artificial overlords. 

The Evolving Artificial Organism

A taste of things to come, researchers are finally firing-up an artificial organism to record how it evolves from primitive unicellular origins to hyper-plexed associative memory network.

The artificial organism unfolds in a virtual world at over ten-thousand generations per hour (kind of hard to do in real life). We can then see how higher beings develop the ability to create associations, and eventually use this knowledge to build more intelligent robots.

Good thing olfaction is the prototypical primordial sensory system, because that’s why this new research is being posted right here. But think about this for a moment – there is no artificial nose. We already have the seeing retina, the hearing cochlea, and even a hand that feels. The nose however, has not been reverse-engineered.

There is a true challenge in replicating the sense of smell, and that is because our sense of smell is programmed by our autobiography. Smells don't mean much to us outside of our subjective experience with them. You just can't upload a dictionary of smells into an electronic nose and expect it to recognize random odors in its environment.

The only way you could do that is if you had a robot that grew up, just like a little kid, with multimodal experiences, social integration, and existential episodes, all associated together and built together into the tangled ball of nerve fibers that we call Self.

Your robot would then have its own limbic system, programmed by a childhood of interaction with the world. It would have to develop a life of its own, an autobiography. This self-identity would then be the substrate upon which the odor network is built. It could then recognize odors, as they would stimulate physiological and emotional responses and associative episodic memories.

Because smell is so tied to our limbic system, it requires a body in order to work. A cerebral organoid isn't a body per se. And neither is an artificially intelligent neural network. And neither is a robot that “comes to life” as a fully-formed adult, all booted-up and ready to go. Humans don’t do it like that. You can’t have a self without a history. (See Patient HM for more on that, however.)

What this new research now reminds us, is that not only does an artificial intelligentity need a body in order to smell, it also needs a lifetime of learning as well.

Notes:

Evolution of learning is key to better artificial intelligence

Sep 2019, phys.org

Anselmo Pontes et al. The Evolutionary Origin of Associative Learning, The American Naturalist (2019). DOI: 10.1086/706252

Post Script:

Finally seeing someone recognize the utility of studying olfaction in the context of machine learning artificial intelligence:

"Srinivasan says he will focus on how noise or variability in odor coding determines the balance between discrimination and learning, explaining that the variability the duo is finding in their work might be a mechanism for distinguishing odors, which could be applied to making better machine learning or AI systems."

How mammals' brains evolved to distinguish odors is nothing to sniff at

July 2019, phys.org