Slashdot Mirror

I would agree with "unexpected," because many do not know that the University of Alberta hosts the Canadian National Institute for Nanotechnology (NINT). Also many are not aware that Robert Wolkow entered the Guinness Book of Records with the "Sharpest Object Ever Made": https://www.ualberta.ca/newtra... But it is the practical translation of this new technology into nanomanufacturing that will make this computer memory revolution possible: https://www.ualberta.ca/scienc... Now, the University of Alberta will not only be know for being the birthplace of Deepmind's AlphaGo, but also for starting the nanomanufacturing revolution.

I would agree with "unexpected," because many do not know that the University of Alberta hosts the Canadian National Institute for Nanotechnology (NINT). Also many are not aware that Robert Wolkow entered the Guinness Book of Records with the "Sharpest Object Ever Made": https://www.ualberta.ca/newtra... But it is the practical translation of this new technology into nanomanufacturing that will make this computer memory revolution possible: https://www.ualberta.ca/scienc... Now, the University of Alberta will not only be know for being the birthplace of Deepmind's AlphaGo, but also for starting the nanomanufacturing revolution.

I disagree. How do you recruit a classical Hebrew scholar to validate your hypothesis and assist with additional work?

You hit up people you know to see if they know any, or anyone who might know any. You ask around the faculty at the university you're associated with. You reach out to other researchers in the same field to see if they know someone or someone who might know someone. You hit Google and find scholars and reach out to them via email. Etc... etc...
 
All of these are professional methods used routinely by serious researchers across any number of fields.
 

You publish your intermediate results and hope that it tickles a suitable person's interest such that they join in the effort.

This is exactly what you don't do unless and until all other approaches have failed to bear fruit. And even then, you plainly mark the results are preliminary and tentative
 
I lack access to the relevant journal, so I have no way to ascertain if they did so. I would not be surprised to find that they did, and the "journalist" that wrote the Gizmodo article or the original University of Alberta press release (that Gizmodo copy-pasted from and failed to credit) simply left it out.
 

You may as will declare Linus' work a joke. It's not as if Linux 0.12 was useful for much. It took a boatload of domain experts to bring it up to the capabilities that made people find it useful.

Um, no. Linux was a part-time non professional project, this linguistics research was (at least in theory) a professional project. The two are in no way comparable.

With the cost of tuition and text books, people should be scamming them.

Have you looked at the cost of tuition in Canada? It is far, far less than the US and now even the UK. At UAlberta the typical total tuition costs (all union, transit etc. fees included) for a Canadian (resident or citizen) student taking a full course load are ~$8k/year for science - and those are Canadian dollars so about US$6k. If you want accommodation and food in a residence the cost rises to just under $16k/year (CAD). You can do the calculation here. The institute in question, Grant McEwan, should be even cheaper. Compare that to the standard £9,000 tuition (~$14,500) in the UK and ~US$40-60k for a top university in the US.

As for the text books, those profits go to the publishers, not the university and frankly the price has started to tick off so many of us faculty that we are either writing our own or using free/open resources at least for lower level courses.

And yet none of them were available to me for the majority of my life. Why is that? It's because nobody had gotten around to the hard work of turning into something actually useful.

I think a study of history would find that that we stand on the shoulders of giants in computing and too often claim old ideas as new. Indeed many useful implementations did exist. We should not use our inability to access something as an excuse to not recognize the amazing contributions of those before us.

Sorry, I wish I could converse more, but spring is coming and I have a lawn to prepare.

History of Parallel Computing https://webdocs.cs.ualberta.ca...

History of Virtualization https://en.wikipedia.org/wiki/...

Mother of All Demos https://en.wikipedia.org/wiki/...

History of the Internet https://en.wikipedia.org/wiki/...

History of Programming Languages https://en.wikipedia.org/wiki/...

Number of papers published and number of papers with at least 1 citation are terrible metrics. There are a large number of conferences and journals where it is trivial to get awful work published. The 1 citation metric also doesn't state that it excludes self-citations. This is on par with stating that something is true because I saw it on wikipedia.

If the numbers hold for conferences on lists such as Conference Rankings and they analyzed impact factor, then that might actually have some merit.

http://poker.srv.ualberta.ca/

The pancreas don't bluff, but sensors do. In the real world, you can't measure everything with perfect certainty, so you have to make guesses. Thing also start to get interesting when you have to make decisions off of these guess, should you gather more information to improve your guess or should you go ahead and make a decision?

One of the people involved also claimed to have solved texas hold em' poker, if this is proven to be true then they've made a major advance in game theory. It means that systems that were too complicated to apply game theory may now be tractable.

they do admit that "It would be disingenuous of us to disguise the fact that the principal motive which prompted the work was the sheer fun of the thing."

This robot kicked MY butt in poker: http://www.newser.com/story/20... http://poker.srv.ualberta.ca/

Imagine you played 4x4 tic-tac-toe but put pictures on the board - one corner is Hitler, another corner is Beyonce, one is the prophet Mohamad, another is Obama. The pictures would likely affect your opponent's play if they don't have the exact optimum strategy memorized. Recognizing their psychological bias would CHANGE your optimum strategy. The psychology would be different playing against an Isis member vs against Jay-Z, so the optimum strategy would be different for each opponent.

I don't think you use the phrase "optimum strategy" the same way I would. While it is true that psychology could infleuence how your oponent might play, if you have the game solved, the oponent's play does not matter. For 4x4 TTT, a winning strategy is outlined here http://all-r-math.blogspot.ca/... which also references Zermelo's theorem showing that "For any finite two-player games of perfect information in which the players move alternatingly and in which chance does not affect the decision making process, one of the players will always have a non-losing strategy. If the game cannot end in a draw, then this non-losing strategy is a winning strategy." Physchology does not enter into it for these types of games.

It seems as though the term "optimum strategy" is being used in the context of "Perfect Play"

https://en.wikipedia.org/wiki/...

But I digress. If in fact the robot is using a "perfect play" strategy, as defined above, then knowing it's strategy won't help. The optimal strategy against "perfect play" is by definition also "perfect play" - any other strategy against perfect play is sub-optimal. It is true (as referenced above) that "perfect play" will never exploit the weaknesses of non-perfect play, but that does not mean that "perfect play" provides any weaknesses that can be exploited by some other strategy.

So, it looks like the site has recovered - have you played yet? Have you been able to consistently win? http://poker.srv.ualberta.ca/

But I don't think you are playing enough hands to really get any useful data. How many hands do you think you need to get data from in order to be able to draw any strong conculsions about the bot's hand based on their bets? And you get no useful data when it folds as I assume you don't get to see their hand then.

And it doesn't really matter anyway - you could know exactly what the bot would do in any situation (by getting a copy of the 11TB of lookup tables), and it doesn't give you any advantage without knowing the hidden cards.

If the authors are correct and they have an optimal playing strategy, the opponent can play any way they want and it doesn't make the robot's job any more difficult - in the long run the optimal strategy will not be beaten by any other strategy.

Of course I could be wrong in my understanding. Do you think your playing is good enough to beat it? Are you planning on giving it a go when the website it not crushed under the weight of everyone else trying to test it out? http://poker.srv.ualberta.ca/

Here is a question: Does it matter if I know the strategy of the opposing player? Looking at the researchers website, the algorithm seems to be deterministic. So I could use the meta-knowledge about how Cepheus would play with any possible hand (that is compatible with my hand and the public cards), and could bet accordingly. From what I've read so far, I don't know if that effect is modelled in the paper.

Poker, for humans, is luck and psychology in the main. It's about convincing the other guy that your hand is better/worse than it is.

But the problem is that if you're playing against a robot, he doesn't care what you want him to think. He knows exactly what the odds are of you having any particular card, and what that means in terms of him beating you. Playing the odds will win on average. It's how it works. It's how casinos get rich enough to have marble floors and air conditioning in the middle of the desert.

If you don't get this, you'll lose a lot of money playing against this machine.

But no doubt you have a "system". Or you can "read" players.

Expert human poker players have a good enough knowledge to know the odds (even if only approximate) on every turn of the card, but they can't analyse every possible combination in time. The rest of it is trying to "lie" to another human. It's rare for a poker player to be the best poker player consistently and for years, precisely because its not as simple as having skill, but overwhelmingly a good amount of luck.

Otherwise, sorry, but anyone on planet would be rich by just plugging in what cards they were given on PartyPoker into an app that tells them the percentage chance of winning. On those kinds of site (last time I used it) there was no human interaction enough to perform any kind of psychology, so it's entirely skill of the game and luck of the cards. And if you can eliminate the need for skill of the game, then by your theory you'd win (almost) every time. You don't. And poker-playing bots only make money when playing against imperfect humans. Play them against each other and you'll be there forever as the money goes back and forth, back and forth (subject to game rules such as blinds, etc.).

Poker has the "most" skill of any casino game. In any of its variants. But that's not a lot. Claiming that a skilled player would beat a bot hands-down? Strange that the poker sites are so hot on blocking bots, then, isn't it?

Bluffing in poker is only relevant in order to make a human opponent make an irrational decision. That's what you're trying to achieve. If the human makes the rational decision every time, then it comes down to luck alone. Making a rational decision every time involves a hell of a lot of card-counting and knowledge of the odds, so few can actually do it properly (I'm a mathematician, I wouldn't dare state that I could calculate the odds without perfect knowledge and a lot of time).

But no amount of bluffing changes the cards in your hand, the cards left in the deck, what the next card will be, or what your opponent probably has in their hand.

The reason people enjoy poker is because a good player can trick a bad player into playing worse. A computer program like this isn't subject to such tricks.

Prove me wrong. Play a statistically-significant number of games against the thing, I believe the link was in the article? http://poker.srv.ualberta.ca/

3x10^14 decisions isn't even in the same range of something like Chess or Go - you can tell this as we can't yet "prove" those games. A decent human can probably play a perfect game. Strange that poker champions tend not to be poker champions for long, unlike Chess champions, Go champions, etc.

More importantly, in Poker, if you can predict what the computer will do, then you can beat it.

If you can say, "oh, the computer just bet $100, that means it thinks it has a 73% chance of winning," you can look at your own hand, and say, "I have an 82% chance of winning, I'll stay in." The computer not only has to be able to bet and fold, it also has to be able to hide its actions from its opponent. Also note that it only calculated the strategy for a two player game, obviously the odds change with more people at the table.

You can actually play against the bot online.

Let me play the computer... I'd bet inside of an hour I can figure out how to make it fold except when it's holding the nuts, and then I'll know it's holding the nuts and it won't be getting anymore money from me. It might win most hands (if it's lucky), but I'll win the most money.

Cepheus calls your bluff.

Just an observation.

If we took the trillion dollars from the Iraq war and invested it in fischer tropsch plants, we could've built 500 fischer tropsch plants which could have converted our natural gas reserves and coal reserves into a whopping 22.5 million barrels of petroleum products a day. With a 40/60 gasoline/diesel split. If I remember correctly, the end price ends up being the equivalent of about 50 dollar a barrel oil. This is roughly equilvalent to the entire oil consumption of the united states.

Source: http://www.ualberta.ca/~tamminga/.hide/10-11/Semester_2/CHE-465/afs_backup/research/Rehan%20Research/42_2_SAN%20FRANCISCO_04-97_0667.pdf
That one is for fischer tropsch processing from natural gas.

Your shredder is too slow? Then you should read slashdot more often! The solution is known!

Studies are done world-wide of lake sedimentology and glacier ice chronology to better understand climatic and human activities over periods where historical records either don't exist or are limited. Have they been done in Peru? Yes. For example, this one deals with human activities over ~1400 years, and this one deals with sediment cores from Lake Titicaca and wet/dry climate cycles. This paper is a good example of how multiple studies from several sites can be combined together to better understand the history of climate change across a larger region (South America and the Carribbean). Lakes in arid areas are particularly interesting because they are so sensitive to changes in climate, so they get studied a lot.

There are additional citations in those papers, and if you search for "Lake Titicaca" "sediment core" and "climate", you'll find plenty of studies in the Peru/Bolivia area, but because most of what you've done is sneer at the idea that anyone could figure out anything about past climates, I doubt you'll really care. However, if other people want to investigate the topic there is plenty available. It is inevitably technical, but the basic principles are not hard to grasp with a bit of effort.

Ok, found it. Neurons operate at 200 Hz, not 10. That gives a brain speed of 24 THz.

University of Alberta aurora watch service:

http://corona-gw.phys.ualberta.ca/AuroraWatch/

On birthday of mine, a few years back, I decided to go to a bar, and I had the pleasure of sitting next to a sociology grad student. The student I sat next to started talking to me about some project he had to do where he had to design a game that encompassed a sociological theme. I happen to have design a few games in my life and so we started digging into the idea. In the end we decided the easiest, and very power, thing that could be done, was to take nearly any existing game, like monopoly, and just given everyone random starting conditions. Use an exponential scale, role two dice and give each person their starting money based on that. Yes it's possibly for the lowest person on the totem pole to succeed and win the game, but chances are that they will not. And this is with only benefit of starting conditions. In reality the benefit of starting conditions is even greater.

In the real world, some pretty innocent starting conditions can make a significant difference in achievement later in life.

G.H. Hardy wrote:

I had better say something here about this question of age, since it is particularly important for mathematicians. No mathematician should ever allow himself to forget that mathematics, more than any other art or science, is a young man's game. To take a simple illustration at a comparatively humble level, the average age of election to the Royal Society is lowest in mathematics. We can naturally find much more striking illustrations. We may consider, for example, the career of a man who was certainly one of the world's three greatest mathematicians. Newton gave up mathe- matics at fifty, and had lost his enthusiasm long before; he had recognized no doubt by the time he was forty that his greatest creative days were over. His greatest idea of all, fluxions and the law of gravitation, came to him about 1666 , when he was twenty- four—'in those days I was in the prime of my age for invention, and minded mathematics and philosophy more than at any time sine'. He made big discoveries until he was nearly forty (the 'elliptic orbit' at thirty-seven), but after that he did little but polish and perfect.
Galois died at twenty-one, Abel at twenty-seven, Ramanujan at thirty-three, Riemann at forty. There have been men who have done great work a good deal later; Gauss's great memoir on differential geometry was published when he was fifty (though he had had the fundamental ideas ten years before). I do not know an instance of a major mathematical advance initiated by a man past fifty. If a man of mature age loses interest in and abandons mathematics, the loss is not likely to be very serious either for mathematics or for himself.

One of the ostensible purposes of such prizes is to subsidize further research. If the recipient of a Fields Medal is past his or her prime, the monies will be wasted, Hardy's observation may no longer hold, but old traditions die hard.

Sprawl allowing more exercise? I wondered WTF you were on about but then I saw "Portland" in your address. If you're in Portland, OR, you certainly see the positive walking experience afforded by decent city planning. I've been there, and plan to move there permanently. I currently live in Memphis, in the "Midtown" area, our most walkable area, which is kinda like Portland in terms of mixing business and residential. Outside of Midtown, the scene is nothing like the non-downtown parts of Portland though... miles of residential with *nothing* else mixed in... not a couple extra blocks to walk to something... think nothing to walk to within an hour or more besides more McMansions. And maybe sidewalk connecting the McMansions in a subdivision, but no sidewalks outside of it leading to the nearest stores or restaurants. Sprawl so far beyond infrastructure that it's rural-style ditches on either side of the road, and lots of narrows that are unsafe to walk or even bike. Our Cordova and Collierville make Gresham and Beaverton look like active-walking-person-paradise. Cordova was farmland in the '90s. Now it's miles of one central road with businesses, and a miles of depth of residential to either side, with hardly any safe walking or biking routes between. The places locals lived before at least had some mixed zones, and things on parallel streets. This has happened all over the US... Portland did some rare forward thinking to stop that crap there, so its worst cases of "sprawl" look like central areas of the cities that grew through the 80's 90's and 00's. Also, scientific studies here and here.

As for what High Fructose Corn Syrup does to appetite: John Hopkins and Iberaki study discussed and linked here. Princeton study here. HFCS also has the business benefits you mentioned... but increasing appetite in your consumer audience is one hell of a business incentive to include it too.

I'd love to see better caloric information, awareness, and data spread... that may help quite a bit. Meanwhile, some things changed in the US a few decades ago and there was an obesity explosion. Some pre-existing, high awareness of caloric intake didn't disappear, but other things certainly happened. If and when there's a serious effort to solve this problem on a wide scale (ahem), that effort needs to include these and/or other well-linked causal factors.

This is old but these types of programs have been out for a long time. I took part in a rock paper scissors AI contest ten years ago.

http://webdocs.cs.ualberta.ca/~darse/rsbpc.html

Beating a human player at rock-paper-scissor is easy. Computers playing against each other is much more fun. There used to be a computer RoShamBo (same game, different name) competition, see: http://webdocs.cs.ualberta.ca/~darse/rsbpc.html

"...it may soon be used as an effective treatment for many forms of cancer."

http://www.dca.med.ualberta.ca/Home/Updates/2007-03-15_Update.cfm

"The University of Alberta Discovery

DCA is an odourless, colourless, inexpensive, relatively non-toxic, small molecule. And researchers at the University of Alberta believe it may soon be used as an effective treatment for many forms of cancer.

Dr. Evangelos Michelakis, a professor at the U of A Department of Medicine, has shown that dichloroacetate (DCA) causes regression in several cancers, including lung, breast, and brain tumors.

Michelakis and his colleagues, including post-doctoral fellow Dr. Sebastien Bonnet, have published the results of their research in the journal Cancer Cell.

Scientists and doctors have used DCA for decades to treat children with inborn errors of metabolism due to mitochondrial diseases. Mitochondria, the energy producing units in cells, have been connected with cancer since the 1930s, when researchers first noticed that these organelles dysfunction when cancer is present.

Until recently, researchers believed that cancer-affected mitochondria are permanently damaged and that this damage is the result, not the cause, of the cancer. But Michelakis, a cardiologist, questioned this belief and began testing DCA, which activates a critical mitochondrial enzyme, as a way to "revive" cancer-affected mitochondria.

The results astounded him.

Michelakis and his colleagues found that DCA normalized the mitochondrial function in many cancers, showing that their function was actively suppressed by the cancer but was not permanently damaged by it.

More importantly, they found that the normalization of mitochondrial function resulted in a significant decrease in tumor growth both in test tubes and in animal models. Also, they noted that DCA, unlike most currently used chemotherapies, did not have any effects on normal, non-cancerous tissues.

"I think DCA can be selective for cancer because it attacks a fundamental process in cancer development that is unique to cancer cells," Michelakis said. "One of the really exciting things about this compound is that it might be able to treat many different forms of cancer".

Another encouraging thing about DCA is that, being so small, it is easily absorbed in the body, and, after oral intake, it can reach areas in the body that other drugs cannot, making it possible to treat brain cancers, for example.

Also, because DCA has been used in both healthy people and sick patients with mitochondrial diseases, researchers already know that it is a relatively non-toxic molecule that can be immediately tested patients with cancer.

"The results are intriguing because they point to the critical role that mitochondria play: they impart a unique trait to cancer cells that can be exploited for cancer therapy"
Dario Alteri
Director University of Massachusetts Cancer Center

Investing in Research

The DCA compound is not patented and not owned by any pharmaceutical company, and, therefore, would likely be an inexpensive drug to administer, says Michelakis, the Canada Research Chair in Pulmonary Hypertension and Director of the Pulmonary Hypertension Program with Capital Health, one of Canada's largest health authorities.

However, as DCA is not patented, Michelakis is concerned that it may be difficult to find funding from private investors to test DCA in clinical trials. He is grateful for the support he has already received from publicly funded agencies, such as the Canadian Institutes for Health Research (CIHR), and he is hopeful such support will continue and allow him to conduct clinical trials of DCA on cancer patients.

Michelakis' research is currently funded by the CIHR, the Canada Foundation for Innovation, the Canada Research Chairs program, and the Alberta Her

Dichloroacetate (DCA) is a cheap, un-patentable, drug (essentially 1vinegar molecule+2chlorine atoms) currently used to treat a rare enzyme disorder in children, but researchers have found it useful in allowing cancer cells to learn how to kill themselves with reasonably acceptable temporary side effects. See "DCA and How It Works" below.

There is almost no funding for this drug study due to it being un-patentable despite quite encouraging results, and reasonably acceptable and reversible side-effects.

Recent human trial reported here:
http://www.medindia.net/news/Dichloroacetate-Effective-Against-Aggressive-Brain-Cancer-68867-1.htm

Initial news from a couple of years ago...

http://www.dca.med.ualberta.ca/Home/index.cfm
http://www.newscientist.com/channel/health/mg19325874.700-cheap-safe-drug-kills-most-cancers.html?DCMP=ILC-Top5&nsref=mg19325874.700

Here's an excerpt...
"DCA and How It Works
Dichloroacetic acid versus Sodium Dichloroacetate

Dichloroacetic acid is a small molecule, basically acetic acid with 2 chlorines. The molecular formula is Cl2CHCOOH.

Dichloroacetate is the sodium salt of dichloroacetic acid. Replace a hydrogen with sodium and you get Cl2CHCOONa

If you view the video from CTV you will see a jar of dichloroacetic acid prominently displayed. http://www.depmed.ualberta.ca/dca/vid1.htm is well worth watching. But they used a “cheap ...powder”. Dichloroacetic acid only comes in liquid. The powder is the sodium salt of dichloroacetic acid. It is sodium dichloroacetate. The researchers did not use the acid.

For those of you searching for DCA, do not buy the acid. I posted info on the FAQ about it. The acid is not the same thing as the acetate. The acid is dangerously corrosive.

How does DCA work, briefly?

The Michelakis team reports that DCA turns on the mitochondria of cancer cells, allowing them to commit cellular suicide, or apoptosis.

Cancer cells shut down the mitochondria, which is the part of the cell that is involved in metabolism and, incidentally, initiates the cell suicide.

A non-cancerous cell will initiate apoptosis when it detects damage within itself that it cannot repair. But a cancer cell resists the suicide process. That is why chemotherapy and radiation treatments do not work very well and actually result in terrible side effects the healthy cells actually die much easier.

Michelakis and his team discovered that they could re-activate the mitochondria of cancer cells. Not only that, the DCA is very effective in doing it: To quote from the Michelakis paper: “The decrease in [Ca2+]i occurs within 5 min and is sustained after 48 hr of DCA exposure.” The mitochondria are so sensitive to DCA that just 5 minutes of exposure reactivates them for 48 hours.

The metabolic approach to cancer is supported by other research. Inhibition of Glycolysis in Cancer Cells: A Novel Strategy to Overcome Drug Resistance Associated with Mitochondrial Respiratory Defect and Hypoxia is a paper by a John Hopkins research team supporting this approach.

http://www.thedcasite.com/dcaforum/DCForumID1/79.html is a post on our chat room by Willis. giving a prediction as to which cancers DCA might not control, and it is being supported by the reports we are receiving."

More on the left side of this web page:
http://www.thedcasite.com/dca_how_it_works.html

= 9J =

Dichloroacetate (DCA) is a cheap, un-patentable, drug (essentially 1vinegar molecule+2chlorine atoms) currently used to treat a rare enzyme disorder in children, but researchers have found it useful in allowing cancer cells to learn how to kill themselves with reasonably acceptable temporary side effects. See "DCA and How It Works" below.

There is almost no funding for this drug study due to it being un-patentable despite quite encouraging results, and reasonably acceptable and reversible side-effects.

Recent human trial reported here:
http://www.medindia.net/news/Dichloroacetate-Effective-Against-Aggressive-Brain-Cancer-68867-1.htm

Initial news from a couple of years ago...

http://www.dca.med.ualberta.ca/Home/index.cfm
http://www.newscientist.com/channel/health/mg19325874.700-cheap-safe-drug-kills-most-cancers.html?DCMP=ILC-Top5&nsref=mg19325874.700

Here's an excerpt...
"DCA and How It Works
Dichloroacetic acid versus Sodium Dichloroacetate

Dichloroacetic acid is a small molecule, basically acetic acid with 2 chlorines. The molecular formula is Cl2CHCOOH.

Dichloroacetate is the sodium salt of dichloroacetic acid. Replace a hydrogen with sodium and you get Cl2CHCOONa

If you view the video from CTV you will see a jar of dichloroacetic acid prominently displayed. http://www.depmed.ualberta.ca/dca/vid1.htm is well worth watching. But they used a “cheap ...powder”. Dichloroacetic acid only comes in liquid. The powder is the sodium salt of dichloroacetic acid. It is sodium dichloroacetate. The researchers did not use the acid.

For those of you searching for DCA, do not buy the acid. I posted info on the FAQ about it. The acid is not the same thing as the acetate. The acid is dangerously corrosive.

How does DCA work, briefly?

The Michelakis team reports that DCA turns on the mitochondria of cancer cells, allowing them to commit cellular suicide, or apoptosis.

Cancer cells shut down the mitochondria, which is the part of the cell that is involved in metabolism and, incidentally, initiates the cell suicide.

A non-cancerous cell will initiate apoptosis when it detects damage within itself that it cannot repair. But a cancer cell resists the suicide process. That is why chemotherapy and radiation treatments do not work very well and actually result in terrible side effects the healthy cells actually die much easier.

Michelakis and his team discovered that they could re-activate the mitochondria of cancer cells. Not only that, the DCA is very effective in doing it: To quote from the Michelakis paper: “The decrease in [Ca2+]i occurs within 5 min and is sustained after 48 hr of DCA exposure.” The mitochondria are so sensitive to DCA that just 5 minutes of exposure reactivates them for 48 hours.

The metabolic approach to cancer is supported by other research. Inhibition of Glycolysis in Cancer Cells: A Novel Strategy to Overcome Drug Resistance Associated with Mitochondrial Respiratory Defect and Hypoxia is a paper by a John Hopkins research team supporting this approach.

http://www.thedcasite.com/dcaforum/DCForumID1/79.html is a post on our chat room by Willis. giving a prediction as to which cancers DCA might not control, and it is being supported by the reports we are receiving."

More on the left side of this web page:
http://www.thedcasite.com/dca_how_it_works.html

= 9J =

Chess has finally been solved to the point that there's now unbeatable AIs available to the average user (assuming it gets to move first)...

No, checkers has been solved to that point. The solution is available online. Perfect play leads to a draw.

Computer chess is merely at the point that if you haven't been on the cover of Chess Life, you're going to get trounced. Even if you have, you're going to lose more than you win. The current situation is that Deep Rybka 2010 has an ELO rating around 3150. That's running on a 4-core AMD-64 desktop machine. The all-time human record is 2851, which Garry Kasparov had in 1999-2000.

The SLOWPOKE Nuclear Reactor Facilityis located on the main campus of the University of Alberta in Edmonton, Alberta, Canada.

To watch the geomagnetic activity live, check AuroraWatch: http://corona-gw.phys.ualberta.ca/AuroraWatch/
You can also subscribe to receive e-mail alerts about probable Northern Lights.
From the website: "AuroraWatch forecasts are made by examining the behaviour of the Earth's magnetic field strength, which is measured by ground-based magnetometers."

In fact, depending on how you define "curiosity", then there are already many examples of programs that are curious.

This is certainly true. reinforcement learning algorithms trade off between exploitation, choosing actions based on the assumption of a static environment, and exploration, testing alternatives, in case the environment has changed. This could be considered a kind of curiosity. What is more interesting to me, as a neuroscientist, is the human ability detect interesting sights or sounds and focus on them. It's like we have a fast but rough novelty detector that can guide our attention towards some event. There is evidence that the amygdala is key element in the neural circuit that detects interesting events, although the mechanism of detection isn't fully understood.

A robot is only as smart as its smartest programmer.

This, under normal defintions of smart, is clearly false. One example: I can program an AI search algorithm to play chess that will make far smarter choices than i would ever be able to (i'm not that good at chess). Some might argue that a search algorithm isn't smart, it's just fast. But to an external observer interacting with the agent, the AI seems much smarter than me, the programmer.

http://www.law.ualberta.ca/centres/ccs/news/?id=332

"The Drug Unit asked Enmax, the local electricity provider, to install a digital recording ammeter (DRA) to record power consumption in Gomboc’s house. Enmax complied without insisting on a warrant. After five days, Enmax gave the police a graph that showed Gomboc’s use of electricity was consistent with running a grow operation. [...] At trial, the Crown conceded that police could not have obtained a search warrant without the data from Enmax."

So, at least in Canada, not only has someone already been subject to a search warrant over electrical usage but the appeals court has ruled it is legal to base a search warrant on someone's electrical usage.

My old university bookstore was one of the first places to get one of these machines, and the sales pitch was that it would be cheaper to get some textbooks from the machine than off of the shelf.

I left before I got a chance to see if that was the case, but people were excited about it.

There is an outreach "master class" scheme involving the LHC where your students can get their hands on data (simulated at the moment but real eventually!). While the tools are simplified compared to what we actually use for an analysis you do get to look at and study real data. You could try talking to CERN to find out if this is available in whatever part of the world you are. We also have a video conference scheme which I've taken part in before where someone from your local university will come and visit and set up a video conference with other people at CERN to discuss the LHC and the physics we do.

There are also various cosmic ray projects that your school can get involved in. If you are in Alberta then your local one is ALTA which is run by a colleague of mine. There are others in various parts of the world as well. These link together multiple schools in a region to build a large air shower array.

Anyone interested in the serious (mathematical, formal proofs, etc.) side of RTS games may find the following an interesting companion read: ORTS. The goal of this project is to develop an RTS platform that can serve as a testbed for real AI research. In other words, a supercomputer could play against you, or even help you by controlling some of your units (roll your own client).

I'm not involved with this project in any way, but it looks pretty exciting. It looks like a bunch of people who contributed to this project wound up at Bioware.

The mathematician's patterns, like the painter's or the poet's must be beautiful; the ideas like the colours or the words, must fit together in a harmonious way. Beauty is the first test: there is no permanent place in the world for ugly mathematics.

There's a link to the paper and some additional information at Robert Wolkow's page.

There's a link to the paper and some additional information at Robert Wolkow's page.

http://www.cs.ualberta.ca/~sutton/book/ebook/the-book.html

These benchmark results are a bit debatable - I've seen different suites electing different "winners" and, while SunSpider seems to be the best, it's a long way from a robust benchmark like SPEC* or DaCapo.

In any event, even if SFX is leading the pack right now, that's because it's the most mature competitor, and its advantage won't last too long. I predict (and I write this logged with my account, not AC, so I would be forever glorified when this becomes true in 12 months max) that both V8 and TraceMonkey will take the lead, leaving SFX in a safe third place permanently.

The reason is very simple. All these new JS VMs are JIT compilers, producing native code. But SFX is a context threaded JIT. Context threading is just a step beyond traditional direct-threaded interpreters: functions are 'compiled' into streams of CALLs into routines that implement each bytecode operation, but there is limited inlining (simple operations and branches), with a focus on reducing branch misprediction.

OTOH, both V8 and TraceMonkey are "real compilers" that emit real native code (not CALL streams) for entire functions (or even larger chunks of code, with inlining). This is necessary to enable traditional optimizations like register allocation, instruction scheduling, constant folding, loop unrolling etc. Some of these optimizations can be performed on a high-level intermediate code representation (HIR), but that's typically not worth the effort without real compilation. E.g., loop unrolling will just waste memory an i-cache efficiency if performed by a threaded interpreter/JIT... as the real benefit of unrolling is giving the compiler a much larger basic block to perform other opts like extra folding and bounds-check elimination, or real low-level tricks like exploring using SIMD registers and operations / Instruction-Level Parallelism / prefetching / branch predication etc.

The only reason why V8 and TraceMonkey don't completely 0wn the benchmarks today, is that these JITs are still in their infancy. They have implemented the foundations (like V8's hidden classes or TM's tracing), but they still miss to implement dozens of important optimizations (including very easy ones - they just didn't have the time yet). Check some comments about V8's limitations. TM's developers have also commented on many limitations, quote (Andreas Gal: "If it talks to the DOM during the benchmark, we currently donâ(TM)t compile across such calls (we plan to for Beta2 though)". This and several other improvements are planned for future builds of Firefox 3.1. Notice that items like special support for DOM interactions and event handlers should be critical to some benchmarks - and of course to real-world RIA apps. I'm sure the V8 hackers are also working around the clock to fill in their own gaps. When both VMs are reasonably mature, SFX will have a VERY hard time competing (unless of course, they abandon the context threading model and mutate into a real compiler). Other optimizations, like JITted regex, can be implemented in all VMs and will eventually be ubiquitous.

Sorry, you are mistaken:
http://sunsite.ualberta.ca/pub/OpenBSD/4.3/i386/install43.iso (203M)

They've had a bootable CD ISO for quite some time, but would be required to do a net install. It's not a big deal since the whole download is just over 100MB. If you couldn't do that you, would need to supply another CD or USB with the install files on it. In the last 2-3 releases, the OpenBSD started created a pre-compiled bootable ISO with all the files included.

It's not a big deal. You can simply extract comp43.tgz from any OpenBSD mirror.. Just for the record, it's 75MB gzip COMPRESSED. But ya, why would you need a compiler for a bootable CD?

Found here: http://sunsite.ualberta.ca/pub/OpenBSD/4.3/i386/comp43.tgz

The history of computer chess is the history of building brute force engines and then refining them by identifying where processing power is successful at winning.

The "brute force engines" are now rather modest. We're way past needing Deep Blue. 4-CPU X86 machines now play at grandmaster level. With 8 CPUs, the computer can spot a grandmaster a pawn and move and still win.

Chess is unforgiving of errors. We now know just how unforgiving. In human grandmaster play, about one move in ten is suboptimal, based on post-game analysis. This is enough to give computers a clear edge. Go isn't that unforgiving.

Checkers, of course, is solved. Perfect play leads to a draw.

Someone needs to define the GP language and fitness function. I'm sorry to say that it's turtles all the way down on this one.

While I'm inclined to agree with you. I wonder if there isn't a fundamental difference between an AI where the computer uses a sophisticated heuristic that was defined by a human, and an GP or reinforcement learning algorithm with a simple reward function.

I put simple reward function in bold, because up until now, (including mogo, i think) there is a lot of "domain knowledge" built into the AI. So indeed, a human harnessing the power of 800-cores beat a human. But in the future, if we really allow a software agent to learn go or any other task by trial and error without specific knowledge (except for of course, the goal of the task) that seems qualitatively different than the current scenario.

And as an amateur go player I should add that a 9-stone handicap is HUGE! (Not as huge as a 25-stone one , of course)

I managed to catch a lecture from the CPRG who are building Polaris 2. While I'm not an AI guy, the gist I got was that there are something like 10^30 states in poker (too many). so what's done here is that there's an abstraction to a simpler game, which is solved, then mapped back to poker.

As well there was a training stage using something called Counterfactual regret... again, Not an AI guy.

All in all, I think the real challenge of computer poker is the incomplete information.

Yeah, but next year will have a 9 after the two 0.

The date is wrong, it's not June, it's July.
http://poker.cs.ualberta.ca/man-machine/

http://poker.cs.ualberta.ca/man-machine/

First match was a draw.

Because as they've said at their page poker has a lot more applications to the real world later. this is all about making intelligent decisions with imperfect information. Chess can simply be brute forced eventually, just like checkers was.

Of course you can, and its one of the more useful features of screen.
MultiUser sessions

You configure ACL's per user, allowing you to give someone read access to your screen but not actually giving them control. This works really well for demonstrations.

It's also a very convenient way to demonstrate to another user on the system some sort of odd behavior you're experiencing, or get feedback on something.

As for the security of it, if security was a serious concern you shouldn't run things in a multiuser system let alone deal with permissions. I trust it, but I also trust every user on the machine. YMMV.