4 ms·
I'll repost my comment from the last version of this article, since I don't think any new developments have invalidated it. This connection to QFT is almost su
by yummyfajitas 16y ago
I'll repost my comment from the last version of this article, since I don't think any new developments have invalidated it.
This connection to QFT is almost surely nonsense. The flag manifold in question comes from matrix groups; basically, consider some subset of 3x3 matrices which satisfy a certain property. I.e., certain types of transformations of a 3d space. While the space of 3x3 matrices has 3x3=9 dimensions, subsets can have fewer dimensions (in this case, 6).
The two-dimensional projections she discovered are functions of the 6 dimensional matrices.
So here is the occams razor explanation for what is happening. Inside the bee's head is an ODE (ordinary differential equation) solver, basically an analog computer. Just think of the ODE solver as a complicated timer, but with output depending on both time and some hidden variable. These are common objects in biology, and don't require many neurons.
The output of the ODE solver is wired to some neurons which reduce 6 variables to 2, according to the formula she discovered.
This is a fantastic discovery, a triumph of applied math. But the connection to QFT is almost surely coincidence: QFT uses symmetries of matrix groups over 3 dimensions, and the bee (which lives in 3 dimensions) also does.
This phenomenon is called universality. Certain objects repeat across diverse areas simply because it is the only logical way (or the most common logical way) that things could happen.
http://news.ycombinator.com/item?id=391396 http://news.ycombinator.com/item?id=391396
- SmokenJoe 16y agoThat made me think more than the article. Why anyone would assume that I find it strange that the articles subject would would be so good at math and yet so willing to disregard Occam's razor. It is a good example of how peoples desires shape their thinking.
- ewjordan 16y agoYou've summed up the technical bit of this perfectly, nothing to add there. Would be a very interesting bit of research, but the speculation really drives me mad, as it's so out there and without evidence to back it up that it's not even worth considering. I have to go further, though - that the researcher is so quick to automatically draw that line connecting two phenomena that both involve group theory as being causally related indicates, to me, that we should be skeptical even about whether the explanation of the bee behavior is correct in any detail. Sometimes when you already know what you really want to "discover", it's very easy to see connections and patterns that don't hold up very well under further scrutiny. The "this group has something to do with bees, and it's also in QFT, ZOMG bees must read quantum fields!" line of logic is so naive (yummyfajitas explained why) that it really makes me suspicious about the whole bit of research. Scientists should keep sloppy thinking like that to themselves (analogy is a great way to guide your research efforts, and once in a while two appearances of the same group in seemingly unrelated phenomena will, in fact, be connected), and when they don't even realize that this kind of thinking is sloppy, it's not a good sign...
- presidentender 16y agoOne could just as well say that thrown objects interact with squares, as a square's area and a thrown object's flight path are both modeled by parabolas.
- swannodette 16y agoAre you a biologist? > Inside the bee's head is an ODE (ordinary differential equation) solver > basically an analog computer. Just think of the ODE solver as a complicated > timer, but with output depending on both time and some hidden variable. > These are common objects in biology, and don't require many neurons. Do you have any sources to back any of this up at all? EDIT: Here's a very long list of papers on Honeybee navigation. http://www.sciencemag.org/content/287/5454/851.abstract http://www.sciencemag.org/content/287/5454/851.abstract. From the looks of no one really knows how it works. That includes the OP.
- jerf 16y agoAn "analog computer" is what a neuron is, or, if you prefer, an analog computing element. And yes, neural nets can do some varieties of them, see http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F4669655%2F4669739%2F04669809.pdf%3Farnumber%3D4669809&authDecision=-203 http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2... for one example (consult Google for many more, '"ordinary differential equation" neuron' just goes on and on) and while that may be computational neurons, real neurons are not less powerful. This is not such an amazing assertion, it's more the sort of thing that is easy to forget non-specialists in the field (or interested laymen) might require citations for.
- swannodette 16y agoWhether artificial neural nets can approximate ODE's and whether neural networks in nature actually do this is not clear. Biology citations please. EDIT: To the reply. None of those papers state that evolution causes naturally occurring neural networks to become ODE solvers.
- yummyfajitas 16y agoI usually hate to argue by authority, but in this case I'll make an exception. Perhaps you are familiar with the work that got Hodgkin and Huxley the 1963 Nobel Prize in Physiology and Medicine? Ok, I won't just be lazy. Here are the original papers: http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1392219&blobtype=pdf http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=139221... http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1392213&blobtype=pdf http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=139221... http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1392209&blobtype=pdf http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=139220... http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1392212&blobtype=pdf http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=139221... http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1392413&blobtype=pdf http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=139241... By the way, the question is not whether ANN's approximate ODEs, but whether ODEs can approximate BNNs (or whatever other processes guide the honeybee dance). [edit in response to your edit: those papers show that a neuron's behavior is well described as an ODE solver. How it evolved is irrelevant.]
- swannodette 16y agoConsider the complexity of bee brains, http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VH9-4292H3S-J&_user=10&_coverDate=02%2F01%2F2001&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1590168478&_rerunOrigin=google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=215f9c81bf69bcc34148d26a3b42a869&searchtype=a http://www.sciencedirect.com/science?_ob=ArticleURL&_udi... The scientist is quite clear that the QFT connection is largely speculative. However, you've hazarded making the same kind of "nonsense" association that you are critiquing regardless of how much more "plausible" or "simpler" you think it is.
- araneae 16y agoAnd from someone who studies neurobiology and animal behavior with a specialty in social insects... This is bullshit. The dance is a solved problem. There are plenty of mysteries about honey bees, and this isn't one of them. A few summers ago I worked with a guy that could read the dance so well that we could physically locate the place the bee was dancing to. (We were studying quorum sensing behavior, which isn't a solved problem; how do bees collaborate to decide which new location to move to?)