Group items tagged

What is the best (wisest) size for a group of individuals? Couzin and Kao put together a series of mathematical models that included correlation and several cues. In one model, for example, a group of animals had to choose between two options-think of two places to find food. But the cues for each choice were not equally reliable, nor were they equally correlated. The scientists found that in these models, a group was more likely to choose the superior option than an individual. Common experience will make us expect that the bigger the group got, the wiser it would become. But they found something very different. Small groups did better than individuals. But bigger groups did not do better than small groups. In fact, they did worse. A group of 5 to 20 individuals made better decisions than an infinitely large crowd. The problem with big groups is this: a faction of the group will follow correlated cues-in other words, the cues that look the same to many individuals. If a correlated cue is misleading, it may cause the whole faction to cast the wrong vote. Couzin and Kao found that this faction can drown out the diversity of information coming from the uncorrelated cue. And this problem only gets worse as the group gets bigger.

Couzin research was the starting point that co-inspired PaGMO from the very beginning. We invited him (and he came) at a formation flying conference for a plenary here in ESTEC. You can see PaGMO as a collective problem solving simulation. In that respect, we learned already that the size of the group and its internal structure (topology) counts and cannot be too large or too random. One of the project the ACT is running (and currently seeking for new ideas/actors) is briefly described here (http://esa.github.io/pygmo/examples/example2.html) and attempts answering the question :"How is collective decision making influenced by the information flow through the group?" by looking at complex simulations of large 'archipelagos'.

The wisdom of crowds breaks down when people are biased. Now researchers have discovered a simple method of removing this bias-just listen to the most confident.

So, those who insist know better? I reckon that an ignorant can also insist on his opinion.

nice paper on crowds .... 

Wisdom of crowds under a new perspective: a motivation for the island model?

Darpa going crowd-sourcing ... for 25 million ....

love this one ... it seems to take physicist to explain to the AI crowd what they are actually doing ... Deep learning is a broad set of techniques that uses multiple layers of representation to automatically learn relevant features directly from structured data. Recently, such techniques have yielded record-breaking results on a diverse set of difficult machine learning tasks in computer vision, speech recognition, and natural language processing. Despite the enormous success of deep learning, relatively little is understood theoretically about why these techniques are so successful at feature learning and compression. Here, we show that deep learning is intimately related to one of the most important and successful techniques in theoretical physics, the renormalization group (RG). RG is an iterative coarse-graining scheme that allows for the extraction of relevant features (i.e. operators) as a physical system is examined at different length scales. We construct an exact mapping from the variational renormalization group, first introduced by Kadanoff, and deep learning architectures based on Restricted Boltzmann Machines (RBMs). We illustrate these ideas using the nearest-neighbor Ising Model in one and two-dimensions. Our results suggests that deep learning algorithms may be employing a generalized RG-like scheme to learn relevant features from data.

Crowd-funded LightSail is sailing in space!

A real scientific study...

We are living in the future when live broadcasts are being censored by AI programs in real-time. I'm sure dictators everywhere are looking forward for these technologies to mature. Having a firewall over reality is so convenient.

Oh, and look - the copyright AI doesn't likes Democrats as well: http://www.wired.com/threatlevel/2012/09/youtube-flags-democrats-convention-video-on-copyright-grounds/

What this tells is that we should not take AI seriously until smart Luis's (or his son) managed to make something decent out of it ... "This was, of course, absurd. First of all, the clips had been provided by the studios to be shown during the award ceremony. The Hugo Awards had explicit permission to broadcast them. But even if they hadn't, it is absolutely fair use to broadcast clips of copyrighted material during an award ceremony. Unfortunately, the digital restriction management (DRM) robots on Ustream had not been programmed with these basic contours of copyright law. And then, it got worse. Amid more cries of dismay on Twitter, Reddit, and elsewhere, the official Worldcon Twitter announced: Chicon 7@chicon_7 We are sorry to report that #Ustream will not resume the video feed. #chicon7 #hugos #worldcon 3 Sep 12 ReplyRetweetFavorite And with that, the broadcast was officially cut off. Dumb robots, programmed to kill any broadcast containing copyrighted material, had destroyed the only live broadcast of the Hugo Awards. Sure, we could read what was happening on Twitter, or get the official winner announcement on the Hugo website, but that is hardly the same. We wanted to see our heroes and friends on that stage, and share the event with them. In the world of science fiction writing, the Hugo Awards are kind of like the Academy Awards. Careers are made; people get dressed up and give speeches; and celebrities rub shoulders with (admittedly geeky) paparazzi. You want to see and hear it if you can. But Ustream's incorrectly programmed copyright enforcement squad had destroyed our only access. It was like a Cory Doctorow story crossed with RoboCop 2, with DRM robots going crazy and shooting indiscriminately into a crowd of perfectly innocent broadcasts."

nice - and we could be the first ones to propose something in the space category ... good ideas welcome! http://www.rockethub.com/projects/scifund/by_category/35-space

what seems to be an impressively detailed space game

> The games we write are always games with a scientific purpose (not training not educational) I'd say investigating how to get the crowd may be an important part of "science of crowdsourcing". So, an obvious example would be comparing how many participants the original ACT space mission game attracted versus a variant implemented in Kerbal and why. Easily made and easily publishable I think. But that's just an obvious example I can give on the spot. I think there is more potential than that, so would not dismiss the idea so definitively. But then, correct me if I'm wrong, social sciences are still not represented in the ACT... Perhaps an idea to revive during the upcoming retreat? ;-)

it's on sale on steam til tomorrow by the way if anyone's interested

Potential lessons for out own trainee program?

well - part of it we already do ... e.g. : (did you see the picture in the report?) "Tolerate a little crowding. It took a little creativity to suddenly find a dozen new workspaces in our two-room office. Fortunately, we've found that a room can always fit one more person-and by induction, you can fit as many as you need. (All those years we spent proving math theorems came in handy after all.) "

talking of interesting business models - > futures markets for whether or not despots will still be in power by the end of the year. Markets>Foreign Affiars/International security

"Challenge the Intrade Crowd with Your Wisdom"... Yeah...

To be published in 2010 by Cambridge Press. Seems quite interesting (small words etc.), there's also a link to download the pre - publication version of 800 pages

Today we released the Astro Drone app. People that have the Parrot AR drone can freely download the game. While they fly their drone in the real world, they are trying to dock to the ISS in the virtual world. But the app is more than a game. Players can choose to participate in a scientific crowd sourcing experiment that aims to improve autonomous capabilities of space probes, such as landing, obstacle avoidance, and docking. If participating, the app extracts visually salient features from the images made by the drone's camera. The features are then combined with the estimates of the drone's state and uploaded. The data is then used in a research aiming to improve robot navigation.

Visit the main ESA website and you'll be greeted with a 6-minute Rosetta promo movie by a kickass Polish artist... P.S. You can also find the video here. P.P.S It seems I've just discovered a way to hijack old diigo entries ;-)