Group items tagged

In a world first, an artificial intelligence machine plays chess by evaluating the board rather than using brute force to work out every possible move. It's been almost 20 years since IBM's Deep Blue supercomputer beat the reigning world chess champion, Gary Kasparov, for the first time under standard tournament rules.

The disadvantage in this kind of engine lies exactly in its inability to extrapolate. You might actually be able to beat it if you play like an idiot.

Cool algorithm! We should implement it in PaGMO.

"We fit data to zonal interaction models and characterize which individual interaction forces suffice to explain observed spatial patterns." You can get the paper from the first author's website: http://people.stfx.ca/rlukeman/research.htm

PNAS? Didnt strike me as sth very new though... We should refer to it in the roots study though: "Social organisms form striking aggregation patterns, displaying cohesion, polarization, and collective intelligence. Determining how they do so in nature is challenging; a plethora of simulation studies displaying life-like swarm behavior lack rigorous comparison with actual data because collecting field data of sufficient quality has been a bottleneck." For roots it is NO bottleneck :) Tobias was right :)

Here they assume all relevant variables influencing behaviour are being observed. Namely, the relative positions and orientations of all ducks in the swarm. So, they make movies of the swarm's movements, process them, and them fit the models to that data. In the roots, though we can observe the complete final structure, or even obtain time-lapse movies showing how that structure came out to be, getting the measurements of all relevant soil variables (nitrogen, phosphorus, ...) throughout the soil, and over time, would be extremely difficult. So I guess a replication of the kind of work they did, but for the roots, would be hard. Nice reference though.

Kind of interesting talk for the "root guys" :)

Be careful, the article is written by the company who did the migration to Ubuntu. Here is a comment by a police guy from IT (in french...). In brief he says that tyhe migration was not a problem for most of the people; exepc for some probleme with access. But it did cost money ! and the saving was not the main argument. "Personnellement concerné par la news qui n'en est pas une, je peux vous assurer que le message de Canonical est surtout commercial... Le choix d'Ubuntu est dû à son hégémonie et le fait que ce soit basé sur du Debian qui est considéré comme très stable. La distrib est d'une maintenance plus aisée que la plupart de celles qui ont été testées. "4500 postes" veut dire "4500 unités de gendarmerie" donc dans les brigades que vous connaissez... Pour ce qui est d'OpenOffice, le passage s'est fait assez tranquillement sauf pour les applis Access qui ont eu un peu de mal à passer sur le module Base...La plupart ont été reprise au sein d'applis php/mysql ou d'applis centralisées... Aujourd'hui, les gendarmes qui je le rappelle ne sont pas informaticiens mais vivent pour vous (au sens le plus strict je vous l'assure) utilisent donc firefox/thunderbird et oppenoffice en clients lourds, le reste étant des applis sur l'intranet ou "invisibles" pour l'utilisateur. Le passage à Ubuntu ne gène en rien dans l'utilisation car le trio précédemment cité est déjà connu et maîtrisé par nombre de mes collègues. Je ne suis pas censé m'exprimer en lieu et place de mes supérieurs mais à titre personnel, le choix d'Ubuntu est un choix intelligent car c'est une distribution avec une prise en main très accessible et avec une maintenance vraiment aisée pour les spécialistes informatiques dont je fais partie...Il ne faut pas oublier qu'une distribution plus élitiste aurait été maîtrisée par moins de monde et donc la maintenance aurait été plus coûteuse... Donc aujourd'hui nous "maîtrisons" cette part de notre infrastructure et la trans

Lotus Notes doesn't run on Linux anyway...

(don't forget to turn your hype-filters on...) MoNETA (http://cns.bu.edu/nl/moneta.html) stands for "MOdular Neural Exploring Traveling Agent". It is one of projects participating in the DARPA-funded SyNAPSE project ("Systems of Neuromorphic Adaptive Plastic Scalable Electronics"): http://www.darpa.mil/dso/thrusts/bio/biologically/synapse/index.htm http://www.darpa.mil/dso/solicitations/baa08-28.html

Nice opinion piece by the creator of C++ Bjarne Stroustrup. Substitute "industry" with "science" and many considerations still apply :)

"for many, "programming" has become a strange combination of unprincipled hacking and invoking other people's libraries (with only the vaguest idea of what's going on). The notions of "maintenance" and "code quality" are typically forgotten or poorly understood. " ... seen so many of those students :( and ad "My suggestion is to define a structure of CS education based on a core plus specializations and application areas", I am not saying the austrian university system is good, but e.g. the CS degrees in Vienna are done like this, there is a core which is the same for everybody 4-5 semester, and then you specialise in e.g. software engineering or computational mgmt and so forth, and then after 2 semester you specialize again into one of I think 7 or 8 master degrees ... It does not make it easy for industry to hire people, as I have noticed, they sometimes really have no clue what the difference between Software Engineering is compared to Computational Intelligence, at least in HR :/

I think it's time to quote "Calvin & Hobbes" (yes, I'm still the guy who knows a C&H for any situation in life) "Sometimes I think the surest sign that intelligent life exists elsewhere in the universe is that none of it has tried to contact us." One should sue this idiot for racism against extraterrestrials!!

"WASHINGTON, D.C. - In a three-month investigation, a team of Northwestern University student reporters has found that the US nation's security establishment is not adequately prepared for many of the environmental changes that are coming faster than predicted and that threaten to reshape demands made on the military and intelligence community. This is despite the fact that the Defense Department has called climate change a potential "accelerant of instability." The Medill School of Journalism graduate student team began publication on January 10 of its findings on the national security implications of climate change with a series of print, video and interactive stories."

I think I start to understand why Italian politics does so horribly bad...

... because they don't follow this rule!

According to the authors we have four types of people in the parlement: 1) intelligent people whose actions produce a gain for both themselves and for other people. 2) helpless/naive people in the top left quadrant whose actions produce a loss for themselves but a gain for others; 3) bandits whose actions produce a gain for themselves but a loss for other people. 4) stupid people in the bottom left quadrant produce a loss for themselves and also for other people. According to the above definition it is clear that their model does not apply to the italian parlament where we only have stupid people and bandits.

A critical view on Prof. Markram's Blue Brain project (in German).

As far as I know it's sort of like "Let's construct an enormous dynamical system and see what happens"... i.e. a waste of taxpayer's money... Able to heal Alzheimer... Yeah... Actually I was on the conference the author is mentioning (FET 2011) and I have seen the presentations of all 6 flagship proposals. Following that I had a discussion with one of my colleagues about the existence of limits of the amount of bullshit politicians are willing to buy from scientists. Will there be a point at which politicians, despite their total ignorance, will realise that scientists simply don't deliver anything they promise? How long will we (scientists) be stuck in the viscous circle of have-to-promise-more-than-predecessors in order to get money? Will we face a situation when we'll be forced to revert to promises which are realistic? To be honest none of the 6 presentations convinced me of their scientific merit (apart from the one on graphene where I have absolutely no expertise to tell). Apparently a huge amount of money is about to be wasted.

It's not just "Let's construct an enormous dynamical system and see what happens", it's worse! Also the simulation of the cosmological evolution is/was a little bit of this type, still the results are very interesting and useful. Why? Neither the whole cosmos nor the human brain at the level of single neurons can be modelled on a computer, that would last aeons on a "yet-to-be-invented-extra-super-computer". Thus one has to make assumptions and simplifications. In cosmology we have working theories of gravitation, thermodynamics, electrodynamics etc. at hand; starting from these theories we can make reasonable assumptions and (more or less) justified simplifications. The result is valuable since it provides insight into a complex system under given, explicit and understood assumptions. Nothing similar seems to exist in neuroscience. There is no theory of the human brain and apparently nobody has the slightest idea which simplifications can be made without harm. Of course, Mr. Markram remains completely unaffected of ''details'' like this. Finally, Marek, money is not wasted, we ''build networks of excellence'' and ''select the brightest of the brightest'' to make them study and work at our ''elite institutions'' :-). I lively remember the stage of one of these "bestofthebest" from Ivy League at the ACT...

Another futuristic view on flying by Airbus. To make it more sexy it also includes some biomimetics...

"The cabin electrical system can be compared to the human brain, with a network of intelligence pulsating through the cabin.". Oh, come on, ...

Tails trying to save their life - Even after they're no longer connected to a lizard brain, gecko tails can flip, jump and lunge in response to their environment.

Single celled organisms may not be able to think, but they can act in remarkable ways

I know, it is not a new concept, but apparently the guys from BMW put some money into it to make is series-proof. Could we - in principle - put something like that onto a spacecraft to solve thermal and energy problems at the same time?