Group items tagged

The best post of Slashdot: "Genetic Algorithms are like the AI equivalent of text editors... everybody has spent a weekend writing one at some point."

Does this have implications for AI algorithms??

I was actually thinking exactly about this question during my bike ride this morning. I am surprised that some codons would need to have a double meaning though because there is already a surplus of codons to translate into just 20-22 proteins (depending on organism). So there should be about 44 codons left to prevent translation errors and in addition regulate gene expression. If - as the article suggests - a single codon can take a dual role, does it so in different situations (needing some other regulator do discern those)? Or does it just perform two functions that always need to happen simultaneously? I tried to learn more from the underlying paper: https://www.sciencemag.org/content/342/6164/1367.full.pdf All I got from that was a headache. :-\

Probably both. Likely a consequence of energy preservation during translation. If you can do the same thing with less genes you save up on the effort required to reproduce. Also I suspect it has something to do with modularity. It makes sense that the gene regulating for "foot" cells also trigger the genes that generate "toe" cells for example. No point in having an extra if statement.

We made it to Slashdot's frontpage !!! :)

Congratulations, gentlemen!

Might be a useful reference for PyGMO-related works.

This is related to the "Verified Software" item in NewScientist's list of ideas that will change science. At the link below you'll find the text of the patents being auctioned: http://icapoceantomo.com/item-for-sale/exclusive-license-related-improved-methodology-formally-developing-control-systems :) Patent #7,627,538 ("Swarm autonomic agents with self-destruct capability") makes for quite an interesting read: "This invention relates generally to artificial intelligence and, more particularly, to architecture for collective interactions between autonomous entities." "In some embodiments, an evolvable synthetic neural system is operably coupled to one or more evolvable synthetic neural systems in a hierarchy." "In yet another aspect, an autonomous nanotechnology swarm may comprise a plurality of workers composed of self-similar autonomic components that are arranged to perform individual tasks in furtherance of a desired objective." "In still yet another aspect, a process to construct an environment to satisfy increasingly demanding external requirements may include instantiating an embryonic evolvable neural interface and evolving the embryonic evolvable neural interface towards complex complete connectivity." "In some embodiments, NBF 500 also includes genetic algorithms (GA) 504 at each interface between autonomic components. The GAs 504 may modify the intra-ENI 202 to satisfy requirements of the SALs 502 during learning, task execution or impairment of other subsystems."

Genetic algorithms are a competitive alternative for training deep neural networks for reinforcement learning. For paper see: https://arxiv.org/pdf/1712.06567.pdf

Interesting pointers in this one! I would like to explore neuroevolution as well, although it seems extremely resource-demanding?

Not necessarily, I think it can be made to be much faster hybridizing it with backprop and Taylor maps. Its one ideas in the closet we still have not explored (Differential Intelligence: accelerating neuroevolution).

Following up on our coffee-time discussion, here's an Evolutionary Algorithm where you are the fitness function, and evolution is guided by your subjective artistic sense. Start from scratch, or pick an existing image in the database, and start evolving. At every generation, you are presented with the individuals/images in the population. Pick the ones you like. Those will be the parents from which the next generation will be bred. Repeat, repeat... where do you get to? If you want to learn more about the science behind this, check the tutorial below by Kenneth Stanley, who is also this site's supervisor: http://dx.doi.org/10.1145/1830761.1830920

The latest from Schmidt / Lipson / Eureqa. A significant improvement over their previous work is that now "The algorithm selects between multiple candidate models by designing experiments to make their predictions disagree."

A simple cube open at the top can increase the annual energy density generation by a factor (depending on the latitude) of 2 - 3.8 compared to a flat horizontal panel, versus an increase by a factor of 1.3 - 1.8 achieved from a flat panel using dual-axis tracking. Genetic algorithm are used to optimize the energy production in a day for arbitrarily shaped 3D solar cells confined to a given area footprint and total volume doi:10.1063/1.3308490 could be interesting to investigate

Nice video featuring the technology. Plus it comes with a good soundtrack! Google's project wing uses a lifting wing concept (more fuel efficient than normal airplane layouts and MUCH more efficient than quadrocopters) but it equips the plane with engines strong enough to hover in a nose up position, allowing vertical landing and takeoff. For the delivery of packages the drone does not even need to land - it can lower them on a wire - much like the skycrane concept used to deliver the Curiosity rover on Mars. Not sure if the skycrane is really necessary but it is certainly cool. Anyways, the video is great for its soundtrack alone! ;-P

Fair enough, but even if you consider advanced/robust/efficient AI, why would you use a drone? Do we envision hundreds of drones above our heads in the street instead of UPS vans, or postmen, considering delivers letters might be more easily achievable. I am not so sure if personal delivery will take this route. On the other hand, if the system would work smoothly, I can image that I'm send a mail with the question whether I'm home (or they might know already from my personal GPS tracker) and then notify me that they are launching my DVD and it will come crashing into my door in 5min.

I'm more curios how they're planning to keep people from stealing the drones. I could do with a drone army myself and having cheap amazon or google drones flying about sounds like a decent source.

...and this is what comes out of this combination.

Octopus and optoelectronics camouflage, light bugs and LEDs, or spider webs and touch screens, ... a whole cool bunch of biomimetic stuff

See also the referred work "Light-extraction enhancement for light-emitting diodes: a firefly-inspired structure refined by the genetic algorithm" - quite cool! https://pure.fundp.ac.be/portal/files/11946897/paper89.pdf