Group items tagged

"Kurzweilians"... LOL. Still not sold on embodiment, btw.

The biggest problem with embodiment is that, since the passive walkers (with which it all started), it hasn't delivered anything really interesting...

The problem with embodiment is that it's done wrong. Embodiment needs to be treated like big data. More sensors, more data, more processing. Just putting a computer in a robot with a camera and microphone is not embodiment.

I like how he attacks Moore's Law. It always looks a bit naive to me if people start to (ab)use it to make their point. No strong opinion about embodiment.

@Paul: How would embodiment be done RIGHT?

Embodiment has some obvious advantages. For example, in the vision domain many hard problems become easy when you have a body with which you can take actions (like looking at an object you don't immediately recognize from a different angle) - a point already made by researchers such as Aloimonos.and Ballard in the end 80s / beginning 90s. However, embodiment goes further than gathering information and "mental" recognition. In this respect, the evolutionary robotics work by for example Beer is interesting, where an agent discriminates between diamonds and circles by avoiding one and catching the other, without there being a clear "moment" in which the recognition takes place. "Recognition" is a behavioral property there, for which embodiment is obviously important. With embodiment the effort for recognizing an object behaviorally can be divided between the brain and the body, resulting in less computation for the brain. Also the article "Behavioural Categorisation: Behaviour makes up for bad vision" is interesting in this respect. In the field of embodied cognitive science, some say that recognition is constituted by the activation of sensorimotor correlations. I wonder to which extent this is true, and if it is valid for extremely simple creatures to more advanced ones, but it is an interesting idea nonetheless. This being said, if "embodiment" implies having a physical body, then I would argue that it is not a necessary requirement for intelligence. "Situatedness", being able to take (virtual or real) "actions" that influence the "inputs", may be.

@Paul While I completely agree about the "embodiment done wrong" (or at least "not exactly correct") part, what you say goes exactly against one of the major claims which are connected with the notion of embodiment (google for "representational bottleneck"). The fact is your brain does *not* have resources to deal with big data. The idea therefore is that it is the body what helps to deal with what to a computer scientist appears like "big data". Understanding how this happens is key. Whether it is the problem of scale or of actually understanding what happens should be quite conclusively shown by the outcomes of the Blue Brain project.

Wouldn't one expect that to produce consciousness (even in a lower form) an approach resembling that of nature would be essential? All animals grow from a very simple initial state (just a few cells) and have only a very limited number of sensors AND processing units. This would allow for a fairly simple way to create simple neural networks and to start up stable neural excitation patterns. Over time as complexity of the body (sensors, processors, actuators) increases the system should be able to adapt in a continuous manner and increase its degree of self-awareness and consciousness. On the other hand, building a simulated brain that resembles (parts of) the human one in its final state seems to me like taking a person who is just dead and trying to restart the brain by means of electric shocks.

Actually on a neuronal level all information gets processed. Not all of it makes it into "conscious" processing or attention. Whatever makes it into conscious processing is a highly reduced representation of the data you get. However that doesn't get lost. Basic, low processed data forms the basis of proprioception and reflexes. Every step you take is a macro command your brain issues to the intricate sensory-motor system that puts your legs in motion by actuating every muscle and correcting every step deviation from its desired trajectory using the complicated system of nerve endings and motor commands. Reflexes which were build over the years, as those massive amounts of data slowly get integrated into the nervous system and the the incipient parts of the brain. But without all those sensors scattered throughout the body, all the little inputs in massive amounts that slowly get filtered through, you would not be able to experience your body, and experience the world. Every concept that you conjure up from your mind is a sort of loose association of your sensorimotor input. How can a robot understand the concept of a strawberry if all it can perceive of it is its shape and color and maybe the sound that it makes as it gets squished? How can you understand the "abstract" notion of strawberry without the incredibly sensible tactile feel, without the act of ripping off the stem, without the motor action of taking it to our mouths, without its texture and taste? When we as humans summon the strawberry thought, all of these concepts and ideas converge (distributed throughout the neurons in our minds) to form this abstract concept formed out of all of these many many correlations. A robot with no touch, no taste, no delicate articulate motions, no "serious" way to interact with and perceive its environment, no massive flow of information from which to chose and and reduce, will never attain human level intelligence. That's point 1. Point 2 is that mere pattern recogn

could we just use genetic algorithms to evolve these shapes and layouts? :P

> Not sure if the skycrane is really necessary but it is certainly cool. I think apart from coolness using a skycrane helps keep the rotating knives away from the recipient...

Honest question, are we ever going to see this in practice? I mean besides some niche application somewhere, isn't it fundamentally flawed or do I need to keep my window opened on the 3rd floor without a balcony when I ordered something from DX? Its pretty cool yes, but practical?

Package delivery is indeed more complicated than it may seem at first sight, although solutions are possible for instance by restricting delivery to distribution centers. What we really need of course is some really efficient and robust AI to navigate without any problems in urban areas : ) The hybrid is interesting since it combines the advantage of a Vertical Takeoff and Landing (and hover), and a wing for more efficient forward flight. Challenges lie in the control of the vehicle under any angle and all that this entails also for higher levels of control. Our lab has first used this concept a few years ago for the DARPA UAVforge challenge, and we had two hybrids in our entry last year for the IMAV 2013 (for some shaky images: https://www.youtube.com/watch?v=Z7XgRK7pMoU ).

Any viral licence is a bad license...

I'm pretty confident I'm about to open a can of worms, but mind explaining why? :)

I am less worried about the can of worms ... actually eager to open it ... so why????

Well, the topic GPL vs other open-source licenses (e.g., BSD, MIT, etc.) is old as the internet and it has provided material for long and glorious flame wars. The executive summary is that the GPL license (the one used by Linux) is a license which imposes some restrictions on the way you are allowed to (re)use the code. Specifically, if you re-use or modify GPL code and re-distribute it, you are required to make it available again under the GPL license. It is called "viral" because once you use a bit of GPL code, you are required to make the whole application GPL - so in this sense GPL code replicates like a virus. On the other side of the spectrum, there are the so-called BSD-like licenses which have more relaxed requirements. Usually, the only obligation they impose is to acknowledge somewhere (e.g., in a README file) that you have used some BSD code and who wrote it (this is called "attribution clause"), but they do not require to re-distribute the whole application under the same license. GPL critics usually claim that the license is not really "free" because it does not allow you to do whatever you want with the code without restrictions. GPL proponents claim that the requirements imposed by the GPL are necessary to safeguard the freedom of the code, in order to avoid being able to re-use GPL code without giving anything back to the community (which the BSD license allow: early versions of Microsoft Windows, for instance, had the networking code basically copy-pasted from BSD-licensed versions of Unix). In my opinion (and this point is often brought up in the debates) the division pro/against GPL mirrors somehow the division between anti/pro anarchism. Anarchists claim that the only way to be really free is the absence of laws, while non-anarchist maintain that the only practical way to be free is to have laws (which by definition limit certain freedoms). So you can see how the topic can quickly become inflammatory :) GPL at the current time is used by aro

whoa, the comment got cut off. Anyway, I was just saying that at the present time the GPL license is used by around 65% of open source projects, including the Linux kernel, KDE, Samba, GCC, all the GNU utils, etc. The topic is much deeper than this brief summary, so if you are interested in it, Leopold, we can discuss it at length in another place.

Thanks for the record long comment - am sure that this is longest ever made to an ACT diigo post! On the topic, I would rather lean for the GPL license (which I also advocated for the Marek viewer programme we put on source forge btw), mainly because I don't trust that open source is by nature delivering a better product and thus will prevail but I still would like to succeed, which I am not sure it would if there were mainly BSD like licenses around. ... but clearly, this is an outsider talking :-)

The good thing about using CdSe NW (used here) is that they show a photon-to-current efficiency window around the visible wavelengths, therefore any visible light can in principle be used in this application to switch the transistor on/off. I don't agree with the moto "Nanowires are also simpler than field effect transistors and so they're potentially cheaper and easier to make." Yes, they are simple, yet for applications, fabricating devices with them consistently is very challenging (being the research effort not cheap at all..) and asks for improvements and breakthroughs in the fabrication process.