Catastrophic forgetting

First days

2011-09-16T12:46:00.001+02:00

It was in Paris that you started dividing, again and again, harnessing the power of exponentials. It is the way of our species, Clyde, and you are the heir to the highest form of life on the planet, the head of humanity’s spear thrown into time. Your mother first felt what the pregnancy tests were later to confirm, what we had wished for in many of our dreams.

From this moment on we marveled over you, at our friends’ place in Washington, our eyes peering beyond a windowless room to catch a glimpse of a future with you, in New-York, as your laughter reached back at us from the steep slopes of time and met playfully with the sounds of the city. In snow-coated Paris, in proud Marseille. Daydreams and smiling absences, brushing out fears, speculating happily at coffee tables. These were the first days of your life as we lived through them. Already you had all our hearts, and we had yet to meet you..

Letters ex Utero

2011-04-03T22:22:00.004+02:00

Dear Baby, I’m just writing to check that everything’s ok, and that you’re comfortable and eat well. Can you sleep? I know it’s noisy in there and you have a lot of work to do so I guess things are not always easy. Your mother caught a cold recently and you must have felt some of the coughing, but she’s been very careful and treated it at once. This is almost over by now, and with the sunny days ahead I think a cold is unlikely to come back. We know you’re doing very well anyway because we can kind of see you with sonograms and we do lots of indirect measures and your mother’s belly has been nicely expanding. But it would be great to have direct confirmation from you when you find the time. Your birth will be the coronation of this summer, let me know if I can do anything to help.

Dear Daddy, I’m sorry I didn’t write before. At first it was mostly because I lacked a nervous system, and now it’s because I’m very busy building one. You’re right that it’s a lot of work, I mean at times I’m building more than 250 000 neurons a minute - yes that’s a quarter of a million! But please know that everything is fine here and Maman has designed a great place for me.

From the very early days I decided to send some of my dividing cells in a delegation, to ask if they could connect to Maman. She has agreed, and our cells merged into a neat little thing through which she’s transiting food and oxygen in just the right proportions. Apparently you people of the outside world strangely insist on eating only a few times in the day, but this is really not the way we do it in here, so we’re very proud of this gadget since it actually regulates the flow. It’s also a filter that keeps harmful things away, you know, like the virus that Mamam has caught. I did feel her cough though, and I hope she’ll get better soon.

Oh also before I forget, it’s always nice when you talk to me, but you should know that I can’t really distinguish the sounds from the background noise of the intestines –sorry, these are sooo loud! However I do feel some pressures when you pass your hand over Maman’s belly.

Well that is, when I’m not sleeping. It’s hard to tell without a clock or anything, but I think I’m sleeping at least one fourth of the time. This week was a tough one (although nothing like the first weeks of course). I managed to put my ears at the appropriate location, and to make some real progress with my genitals, which are nearly completed (I promise that next time you try a sono I won't be sitting on my knees, so as to let you see). I’m also beginning to warp my nervous fibers in myelin since I’m told it helps neurons to talk faster. This is quite an assignment and I gather it will take me many months. But I’m not complaining, the place is great and Maman feeds me well.

Talk soon !

Of sentient beings and kitten pee.

2011-02-26T22:16:00.006+01:00

Sometimes a man must step back for a moment, take a deep breath, and start asking profound questions in all seriousness. This is not the time. Nevertheless...

Why are computers not sentient yet?

Computers have not been around for very long (although perhaps longer than one might think). They are now, however, ubiquitous and clearly here to stay. It has something to do with their universality, which is also why these things are getting more and more powerful at every instant. Given how powerful they already are, in a sense it's flattering that no machine -that we know of- has yet reached consciousness. But how long will this last? How will we know when a machine really becomes conscious? What does this word even mean? How will this new consciousness arise? And what will happen next?

The usual answers to these five questions are, in order:
1. not long;
2. with a Turing test or a variant of it;
3. we don't know but we don't need to know for the Turing test;
4. we have no idea;
5. something big, hopefully nothing bad.

On the last two questions however, screenwriters are actually not short of ideas.

My personal all time favourite scenario would be from a movie in the late 80s, "Electric Dreams", where a commodore computer percolates to consciousness after being spilt some champagne on its keyboard. What happens then? I can't recall exactly but eventually the owner gets laid. Well loads of movies share this simple plot of a machine that has become or is becoming sentient. We have of course Terminator's Skynet, 2001 Space Odyssey's HAL (which by the way is IBM shifted by 1 letter..), or obviously movies like I, Robot..

As for books, it seems you simply can't be a science fiction writter if you don't have your own story. Probably the best I've read is from Dan Simmon's Hyperion, where the so-called "technocenter" has evolved from the very real and ongoing Tierra Project in the artificial life community. I like this scenario because we don't get to design the conscious program architecture, and it doesn't occur by accident either. Rather, consiousness arises out of evolutionary dynamics, it creeps out of the cognitive night by gradual changes and mutations, each of which is beneficial to the artificial species. But it doesn't tell us what the architecture is, or what the critical mutation was, if there was only one.

In all these cases -with the honourable exception of Electric Dreams- we notice that the birth of a conscious machine is always a threat to humanity - while in the superior Electric Dreams it is only a threat to abstinence. It turns out that this concern is shared by some people working in a field called "Friendly AI" (friendly as in "how do we make this conscious machine friendly", because No, Asimov's famous three laws probably won't fly).(note that abstinence might be another shared concern in the friendly AI community).

According to friendly AI, the problem is not really couched in terms of consciousness, but rather in the ability to improve one's cognitive algorithms. Essentially, the idea goes like that: the minute we create a machine that is able to learn how to improve on its own programs, the thing will grow out of control exponentially fast. Before we know, given its presumaby considerable computational resources, it would reach IQs that no human has even contemplated. This moment in our future has been called by some, rather pompously, the "Singularity". It has a leader, an institute, and generally makes many people think hard in the silicon valley, and at least one person in Australia - a very smart and distinguished researcher on consciousness.

Can someone please tell these guys that the probem has been solved in the late 80s by a bottle of champagne and a commodore computer? Admittedly, any kind of liquid could work. In fact in my own estimates kitten pee would be sufficient for the powerful computers we have nowadays. And you even get a chance to rediscover the Welsh language. Genius? Yes.

Studies show that kitten pee might be sufficient to make current laptops conscious, with possibly happy side effects that bear on celtic languages. Courtesy of Dr Kim.

Blue eyes, light and Octopuses

2011-02-06T00:20:00.014+01:00

Some remarkable facts about the eye:

1. until 10 000 years ago there almost certainly was no blue-eyed person on the face of the earth.

This blows my mind a bit..Language and culture are both dated back to something like 100 000 or 50 000 years ago (See J. Diamond's excellent book). So for tens of thousands of years, all these people were doing what humans are known for doing so well, which is mostly having sex, fighting and talking (not necessarily in that order), fighting again, drinking, eating, talking again, making tools and art, playing chance games and having sex - and all of them had brown eyes. Mr check-this-out-chicks-I've-got-blue-eyes must have had an incredible success. No doubt he was perceived as marked by the gods. But did it make him more, or less popular? Well you say, we know the genes have spread, so after all the pooh-pooh couldn't have been that bad.

Now it turns out that I have blue eyes, and so has a friend of mine. That's not uncommon, but since we have evidence that all blue-eyed people can be traced back to Mr check-this-out who lived 10 000 years ago next to the black sea, it follows that our common ancester cannot be older than him. And we're not more than 700 generations appart (to pick an upper bound).

2. The human eye (the so-called "camera eye") has evolved between 50 to 100 times independently in other species..

Combined with a demonstration of how it has evolved in humans, this is a wonderfully simple rebuttal of creationists' "look how complex this is, how could it have possibly evolved?" favourite sentence. Well, it evolved in small steps, that were selected because they provided small advantages, and we're nothing exceptional it happened not one but many times. Another hit in the long series of demotions being forced upon the human race.

But there is still a sense in which we are exceptional. Turns out we humans, like all vertebrates, have our optic fibers wired in the wrong direction. Imagine you just came home, and you turn the lights on. But your light switch has its electric wires coming out of the switch, rather than being hidden behind the switch and running inside the wall like in any other reasonable place.

"Thanks for your work, yes, we will call you back. Thank you."

Well the appartment is still ok, I mean the lights work, you can stil hit the switch on and off, but all these wires coming out and dangling down - it's a little bit peculiar. The guy who did this job, well you probaby won't call him back. Unless, of course you're the eccentric who did it.

As a matter of fact, let's assume even more eccentricity: for some reason let us imagine that all your light switches must be packed on the same wall with their wires dangling out, and the wires must somehow find a way to run behind the wall (I know, who would do that, but bear with me. Imagine you're some kind of a contemporary artist, you know, like in another room you have a painting entirely blue that you call art.). Well I guess you would have no choice: there's one place on the wall where you cannot put any light switch because you must drill a big hole there and feed in all the wires from all the light switches.

That's our blind spot: at this place the human eye cannot see anything because there's no retinal cell, only optic fibers bundling and running to the cortex. So what the cortex does is that it interpolates the stimlus that was likely to be there given the surrounding. If the surrounding region was white, the blind spot is filled with white.

Blind spot demo. Click on the image to make it full size. Close the left eye and fixate the cross. If you're not an octopus and if you slowly move the screen back, or forth, at some point the cow will disappear.

Octopuses did it better - their optic fibers are wired ouside-in, not inside-out-and-in-again.. These guys don't have any blind spot, in addition to being awesome at hiding (i would pay big money for this camo trick, but I don't think we have any real idea how they can do that).

3. Until the 19th century, nobody (who lived to tell) had ever travelled at more than 100 km/h.

For this reason, it is unclear to me why quantum mechanics should be less intuitive than special relativity (where thinking about light speed is critical). After all these are speed regimes and space scales that are both entirely alien to us - we did not evolve for them. So should they not be equally unintuitive? Yet I find it possible to grasp the Lorentz transform, I mean that's not completely beyond me with some practice, but after years the double slit experiment still baffles me (see lecture 6).

Life in Aix

2010-05-28T01:17:00.009+02:00

Two things one should know about Aix.

1. Aix rules.

2. Transports to Aix don't.

Let me explain : I currently live in Aix but work in Marseille, which is 30 minutes away by train. It has been brought to my attention that Marseille has several pretty cool beaches, which is entirely beyond the distractions Aix has to offer - a point that has been made to its inhabitants repeatedly. This is mostly because Aix fails to be situated on anything like a see-front, and appears to have very little intention to take action in that matter.

Recognizing this deep truth, yesterday evening after work I went swimming to the beach "du Prado" with a lab colleague. I found a semi-closed space to change and swiftly dived into our good old flat mediterranean sea, in so doing probably raising the sea level by a few molecules per square kilometers (anyone to compute this? I can't be much more than 80 liters). People of all sea fronts, you know this privileged sensation. The feel of the sea on your skin after a hard day's work, when all stress is lifted up by this primordial backwash that seems to have a private line with every single cell in your body.

The rest of the evening was spent in quite agreeable discussions on social cognition while sharing mussels in a restaurant on the promenade.

And then the true challenge inevitably came up: going back to Aix. When I got to Marseille's railway station I had missed the previous train by a good 20 minutes and the next was only 40 minutes later, that's how frequent trains to this pearl of a city are. In addition the train is an omnibus that would take us there in another 45 minutes if all went well. It was 10pm. I arrived in Aix at 11:37 but not before a homeless woman sat next to me and asked if she could play with my penis (what is wrong with you? I said no.).

But tonight I decided to repeat the operation, and went jogging with another colleague in Marseille after work. We are both post-docs in computational modelling; which makes things simple and brings deep conversations very rapidly. After scaring a few roosters in the parc du 26ème centenaire (yes that's how simple things got), we headed back to Notre Dame du Mont, where my colleague cooked us some truly wonderful dishes (see his co-blog "free your food"). We were still discussing aging processes, physics and epistemology when I figured I had to head back to the station or I'd miss the train.

Did I mention there was a strike on the line this morning and my direct train was cancelled? What you say, a strike in France? Well the strike wasn't over, and the rare trains that were not suppressed had been replaced by coaches. Of course the coach made numerous stops until it eventually reached Aix at something like 11:05pm, again after these 45 seemingly incompressible minutes of transportation, with loud music and a smell of pot coming from the back.

Of course it was tedious and slightly unnerving. But as always once you make it to Aix, all is forgiven. And if in the morning the sun comes to shine just a little bit on "Cours Mirabeau", then everything is illuminated.

Aix's "Cours Mirabeau"

"...to recreate life out of life"

2010-05-22T01:44:00.026+02:00

The genome is now read AND write.

Just a few years after decoding the human genome, Craig Venter et al are making history again! Today Venter issued a public statement at the JCV Institute annoucing the creation of a cell controlled by a synthetic genome...

With what appears to be a frighteningly common computer the team has designed a chromosome, assembled it using very simple chemicals and inserted it into a bacterium. After a number of unsuccessful attempts, they could eventually watch as the synthetic genome kicked out the host's original genetic material, and reboot the cell which now divides with the new genome.

So this is the first self-replicating organism governed by a synthetic code. For those of you who care about definitions, it's not synthetic life really -it uses living material like the cell's cytoplasm etc. We're not there yet. Moreover, what is really artifical is the method used to assemble the genome, because indeed the genome itself is almost a copy-pasted version of a living organism's genome.

Well almost, but not quite!

And that's the best part: hidden in the genetic code of this synthetic bacterium, the team has "hand-written" three citations. One of these is the James Joyce citation of this post:

"To live, to err, to fall, to triumph, to recreate life out of life."

They also wrote a list of more than 40 researcher's names and...a website address! (I know, you need this website address. I do too, so stay tuned.)

Venter simply inspires me. And he's a very active ecologist whose groundbreaking research could well be our salvation. I can only hope he becomes president someday!

What's Your Volcanic Cloud Story?

2010-05-08T17:04:00.001+02:00

Here's what happened to Yours Truly.

The 1st English/Spanish joint conference on cognition had barely begun in fair Granada when ...

...moronic volcano Eyjasomething starts coughing !
(check this out by the way, a moron but with style)

Next thing we knew the European airspace was essentially down.

250 english people suddenly all thinking the same:

"huh .. I guess I'll just fly to Madrid, that still works, then it shouldn't be too difficult to get to the south of France, and from then take a train all the way up where I'll catch a ferry to mother UK"

miiiippp (buzzer sound):

General railway strike in france -train traffic paralized (now isn't that special!). 250 english people suddenly looking...a bit embarassed.

Well to make a long story short, these 250 people eventually rented several coaches and got back home after a 48h trip... I think "memorable" was the term most commonly used to describe the event.

Me? Well I tried to play it cool. I thought I'd litteraly let the dust settle in Granada (nice city by the way).
I said I tried. But Iberia just wouldn't let me win (ever tried to call Iberia to change a reservation? Don't. They have a nice recorded tape sending you to their website. Go to their website and they give you..their phone number)

==> I had to go to Madrid.
Now I must confess I only knew Madrid from its airport so I had very low expectations. Please don't throw stones. I was so wrong - Madrid rules. I loved it. Great food, very alive, surprisingly walkable.

Still, I had to reach France, so ... dashed to Barcelona by high-speed train -funnily looks a bit like japanese shinkansen inside..).

Barcelona..so much to say. Just try the morning croissants at café Ra to get a bit closer to paradise.

Then find someone with a car who goes to Marseille the next morning at 11am, get to the checkpoint and

VOILA!

..

..

..

..

Except, the car got broke into during the night -2 days to fix.

I'm not making this up.
I looked up to the sky.
It was just 11 am.

Although I couldn't see it, I knew the day would be suffused in moronic ash.

-----------> what's your story?

Update: oh come OOn, it's doing it again!

no comment

2009-12-26T00:19:00.013+01:00

via Today's Big Thing.

Great inventions #1 -the fireplace

2009-12-23T00:03:00.003+01:00

And I don't mean fire, mind you, but literally the fireplace. And please spare me the gospel about how it was the discovery of fire that really started out technology. Fire might have started technology, but I contend fireplaces started civilization!

Taming such a useful element and forcing it to be part of the house, just another place in one's home...that was genius. From this moment on and for centuries, the fireplace has been radiating heat and light throughout dark and cold nights in every home that deserves this name. It is only in our hopelessly sophisticated times that people --including yours truly- accept this miserable thing that is the absence of a fireplace.

People all around the world, brace yourself, it is time to rebel. You lucky souls with an unused chimney, can you try and find the strength to stand up and sweep it! And fetch twigs, branches and logs! The rest of us, poor beggars, let us tear our radiators out, let us find another place where we can watch the dancing flames again, and warm our hearts at the sound of softly creaking wood!

Applied cognitive science

2009-12-21T21:30:00.003+01:00

I'm not sure one can find anything more pressing than to help educate children all around the world -that is of course, once their basic health is guaranteed. We shouldn't leave this to religion, as is too often the case, and we should do it with the full knowledge our modern times have to offer. That means exploiting cognitive science.

In this recent NYT article, get acquainted with children who directly profit from our emerging understanding of the brain. New teaching techniques apparently include several forms of playful enumeration using sensory-motor reinforcement (e.g. seven! can you touch your nose seven times?), supervised training in what we call subitization (the ability to rapidly distinguish between one, two or three objects), as well as number decisions (which is bigger, 5, 7 or 9?) set-up in a fun way.

The human system for number is a fascinating thing. To give you but a simple taste of how strange it is, the evidence is that we represent numbers using a logarithmic scale -our notion of the distance between 10 and 100 being roughly the same as between 1000 and 10000. Athough I confess I haven't read the book I have no hesitation in recommanding The Number Sense (the writer Stanislas Dehaene is a great cognitive scientist, see for instance his page on edge).

The big picture!

2009-12-20T19:26:00.003+01:00

The universe could have been quite boring..

fortunately it seems its weirdness will always exceed our wildest expectations.
(via Backreaction)

Why I shouldn't like Marseille

2009-12-13T20:55:00.000+01:00

- The city's loud and dirty (well what do you expect, it's a port)
- You can spend ages trying to find a place with both an internet connection and a plug
- It is full of excited youngsters dressed in cheap sportswear with the OM colours (local football club)

and yet..

The inner language

2009-12-09T10:14:00.002+01:00

Cognitive science is a pretty large field, but three questions have been driving a lot of the research effort, almost from the start.

What is language ?

First notice that it is a two-faced coin, there’s external language and internal language. Everybody knows what external language is : it is that particular set of sounds and signs whose shapes and sequences convey some conventional meaning. This convention is formalized into rules, exceptions and templates, and examplified in googles of more or less beautiful instances in everyday life.

Internal language on the other hand is how the brain represents the external language environment to itself. It is the set of neural mechanisms and representations by which we humans can write and read, speak and understand speech –that is to say, by which we can in turn contribute to external language. Cognitive science is interested in internal language.

There is a longstanding disagreement among researchers concerning the mere mechanisms and representations of language. The so-called classical school holds that it’s all rules and symbols, in direct legacy of Turing’s universal machine. This school has reached a venerable age but it is alive and very influent, with prominent researchers such as Chomsky, Pinker or Fodor to give a few names, and the next generation lead by very smart people such as e.g. Marcus. In its strongest Fodorian form –admittedly not the most widespread- the classical school adheres to the principle of multiple realizability stating that brain features are utterly irrelevant to understand internal language: the brain is just a transparent implementational device.

The connectionist school of thought starts from the opposite premise : the brain does set constraints on how we implement language. So this school must work within neural-like frameworks, and it studies the behaviour of devices made of numerous parallel and distributed processing units connected to one another. The knowledge of the network is embedded in its architecture and in the strength of its connections. So at bottom, connectionism holds that it’s all about associations and network states. Connectionism has been very seriously criticized all along its history, to the point of being sometimes left for dead, but it has grown stronger from every blow. This tradition has pioneers such as McClelland, Rumelhart, Hinton or Elman; and with a new generation perhaps lead by people like O’Reilly, Seidenberg, Smolensky and Plunkett. Ok, as you'd have guess, I for one am a connectionist.

There are two remaining questions I would like to describe in a another post: Is language innate? And is there a language module ?

What do barnacles, toxoplasmosis and rabies have in common?

2009-12-05T15:55:00.008+01:00

These are all hijacking machines.

So instead of digging its own hole in the sand and lay down its eggs, the female sacculine barnacle finds it much more fun to adhere to a green crab's back, slowly move on the shell's surface until it finds a joint, and then send its cells invading the body.

It develops until it reaches the underside, where it drills an opening for the male sacculine barnacle to enter. The couple of parasites then make thousands of eggs, castrating the crab in the process and rewiring it to their purpose. What if the crab is a male? Well it seems in this case the parasite secretes a substance which demulitplies the crab's estrogen level. Poor crab shortly starts digging a perfectly fine hole -except of course it is not its eggs it is laying. Ah, ..nature's cruelly unsensitive genius..

I you think that wasn't so cute, rabies is worse. As is well known, this lethal virus most commonly propagates through bites (98% of human infection cases come from dog bites). Once in another mammal's body, the virus completely disdaigns any other organ than the brain, towards which it immediately starts migrating. This can take months, but once it's there all bets are off. In a very real sense this organism knows more about the brain than any living neuroscientist. It also "knows" it can only be passed on by making the host bite as many animals as possible (or rather, those ancester rabies viruses whose dna did not program them to act accordingly did not reproduce) . So it makes the host extremely agressive, with a strong need to bite all living things arounds.

But mind you, that's not sufficient: because saliva is essential in the spreading, the virus has to make the host produce a lot of saliva. His way to do that is to migrate to the salivary glands and crazily stimulate them. The host starts producing plenty of saliva. But wait, what if the host drinks and looses this saliva? Well rabies has "thought" it all, and simply makes it impossible for the host to swallow any kind of liquid..the host finally dies paralyzed, or in the rare documented cases of remission, with severe brain damages.

Finally we come to toxoplasmosis..here's an equally impressive, although slightly more subtle virus. Lethal only if you're a rodent (or so it would seem). I highly recommend this Edge video to get an up-to-date, cutting edge understanding of what is going on.

Religion hurts more than it heals -let's get rid of it.

2009-12-04T23:44:00.003+01:00

It hurts. badly. Kills millions, scares more. It is especially hard on women all other the world, and contaminates children shamelessly from a young age. It has infected more than 3/4 of the world's population. I'm not talking about an organic disease, but about religion. This merciless plague is the love of god. Are you religious? Please read more.

If you're a non-practitionner, benevolent, actively helping or enlightened religious person, you might think there's nothing wrong in your faith.

But there is no such thing as a good religion or an enlightened religion: religion cannot be enlightened because it requires one to believe in something in the absence of light (evidence). There is no such thing as a harmless faith in god: everytime we join our hands together and pray for the intercession of a supernatural being we stone a woman to death, we support past, present and future murderous cruisades. I can run a nice, open-minded religious orphanage in France but ultimately if I teach the kids to believe in god, then I precipitate lethal weapons into the hands of other children in Palestine.

Why am I saying this? Because the idea that it is OK to believe in a supernatural being is precisely the one which drives or has driven cruisades, terrorist attacks, the public stoning of adulterous women and all kinds of violences on children. If I help propagate this idea anywhere in the world as part of my good deed I also become an accomplice in this ongoing butchery.

What can I do?

We must help religious people to get out of this illness. We must stop saying "I respect your faith", "I respect your religion" and so on. Would we respect someone's aids, perhaps the second most frightening spreading virus after religion? No, we would help this person in this terrible struggle, and it goes without saying, he or she would refrain from doing anything that could propagate the virus.

So the first thing to do is to point to the body count.
Then, secure the idea not to believe in anything without evidence.
Finally, realize that one can do good things without buying the god package.

Atheism should follow shortly !

UPDATE 13/12/09: now that is just priceless!

Some standard religious misconceptions (and how to expose them)

2009-11-26T19:47:00.003+01:00

You can't disprove god

Ah, that's an easy one, abundantly and eloquently addressed by Dawkins et al. To sum it up:

Most things cannot be diproved, this doesn't make them real.

For instance I cannot disprove toothfairies, this doesn't mean their existence is as likely as their non-existence.
The likelihood for the existence of toothfairies, or any kind of supernatural entity, is equally low and can be safely neglected, given that there isn't a hint of evidence for them.

Big bang is evidence for god

Again, I think the best answer here is Dawkins's one. Debating with a theologian who was thundering that the Genesis had made one stunning prediction: that the universe had been created -that it had a single origin in space-time- Dawkins answered something like:

"Well either it has been created or it has not. Flip a coin and call a side,

you'll be right 50% of the time"

Being at chance level fails to be impressive.
Moreover, I would like to mention that not all researchers buy the big bang theory. The endless universe alternative is alive and supported by foremost physicists.

Fine tuning is evidence for god

Yep, that's a hard one, I think the most difficult. It is reminiscent of creationists argument on the eye complexity. If you can't invoke the theory of evolution, it is very hard to understand how such complexity could have come into existence without a designer. But once evolution is discovered the eye makes perfect sense (just watch this stunning video, hum again by Dawkins, well he's very good at busting these arguments), In fact the eye has been invented independently by dozens of species, and frankly the human one is not so well designed..

So the answer for fine tuning, here again, is very probably evolution. The inflationary scenario seems to come with eternal inflation, which would produce tens of thousands of googles of bubble universes, each of which would have a different tuning. Other researchers have advocated a kind of cosmological natural selection mediated by black-holes -they would be the place where other baby universes are created.

Religion provides moral guidance

Well that is, when it doesn't tell you to kill people. Plenty of things provide moral guidance. Introspection, friendship, benevolent work..The point is:

Why not look for a kind of moral guidance where you actually don't have to buy the

whole "belief-without-evidence" package

From the theist: Quitting my religion would hurt people I love

Well, if you feel like quitting, you're already sick with preaches and cult and nonsensical arguments and bigotry etc.., and you already feel bad because you participate or silently consent to all this sick business.

So first: you should stop hurting yourself and quit.

Then peacefully explain to your religious relatives and friends that you cannot keep on hurting yourself.
People who really love you will understand and accept your new difference. The others..well these were relationships probably not much worth cultivating anyway.

From the atheist: not going to some religious ceremonies would hurt people I love

Idem!

Beauty and design

2009-11-22T21:32:00.005+01:00

Simple idea, great execution.

Also, breathtaking pictures from National Geographic, very nice to look at while listening for instance to Portishead.

(Via Webdesigner depot, where one can find all kinds of interesting things. Here a wonderful pencil sculpture by Jennifer Maestre)

Inspired by all this beauty, my wife and I went on to pull the best late brunch/tea time together EVER !

painful post alert - the Falun Gong genocide

2009-11-21T15:23:00.003+01:00

You're peacefully enjoying your first coffee of the day close to place de la Bastille in Paris. Engaging in idle chit-chat with the waitress, anticipating a very nice and surprisingly warm WE. This is when the trumpets and drums kick in.

Shortly after, a hundred asian people in white uniforms have filled the boulevard. At first you smile, because it seems like a happy procession or something, and they smile too. Then you notice the determined looks, and the second wave comes in with signs and banners. This is not just another procession, it is a hopeful march of protest, a call for help:

"stop the Falun-Gong genocide"

"chinese communist party = 80 000 000 deaths"

Like many, I had heard of Falun Gong before (a.k.a. Falun Dafa), and they had my sympathy. Whether Falun Dafa is or is not a cult is utterly irrelevant. These people are emprisonned and tortured all over china for decades for their beliefs.

The chinese communist party is truly a disgrace to the human race. It has become perhaps the most violent religion ever, overtly murderous and crushing any opposition without any kind of shame. They kill people by thousands while buying the Olympic games from us, with a smile.

But I had never investigated this more deeply -and I hadn't been handed the flier. here's a (heavy) pdf version very much like the one I hold.

As you can see, some aspects of Falun Gong leave us with a bit of a mystical taste, but otherwise this appears to be mostly a harmless set of practical exercises for inner development, with some more or less deep spiritual theory behind. Here Falun Gong is described as a spiritual practice without worship or rituals, which encourages authentical thought, good-willingness and tolerance.

But, prepare yourself for page three.

"nullius in verba", take no one's word for it, so check the veracity of their claims. But this looks to me like a pretty serious investigation. If they are right, then we should all take at least one minute to do something, now. I suggest to start here.

The Scandal of Probabilities?

2009-11-21T01:17:00.002+01:00

According to the late Bertrand Russel, there is a scandal at the root of probability theory: it is the scandal of induction.

Simply stated, the problem is that what has been verified a finite number of times has no reason to be verified beyond these occasions!

Because we can only make a finite number of experiments, we can never really infer from experiment the validity of any statement. Exit the frequentist interpretation of proability theory, which sees the probability for any event as the limit of its frequence. And other interpretations of probability would also appear not to make sense.

Now this is slightly embarassing.

And not the least because our second most successful scientific theory, quantum theory (you can probably guess what the first is?), tells us that our universe is fundamentally probabilistic (Feynman seems to never have fully overcome a certain reluctance for this feature of the theory, a feature he once half-jokingly called an abomination). Well but it's ok, because nobody really understands QT.

Could two scandals cancel each other? Well, some prominent scientists such as David Deutsch think that the scandal of induction is problematic for all but one interpretation of QT, namely the many-world interpretation. That's an idea worth tracking, although the many world interpretation is obviously completely crazy (and thus unsurprisingly, favoured among physicists).

Extirpating religion

2009-11-14T23:34:00.010+01:00

I have already compared religion to tobacco elsewhere, slightly varying on Marx. Having been slave to both earlier in my life (not to opium or Marx, mind you), I feel entitled to comment and actually think the comparison is quite warranted.

They both provide tremendous mental comfort.

I'm not sure whether people who have never smoked realize that lighting up a cigarette everytime one gets upset, tired, annoyed, cold or lonely is just a stupendous support. 5 minutes of distilled pleasure at will. (I personally remember when I quit smoking, the overwhelming feeling was: wait. something's wrong. where's all the daily pleasure gone..where do they find it? Well they don't, and I just went back to walking without crutches)(and took 5 kilos)(and started drinking a little more ^_^).

Well clearly religion works just the same for one's mental discomforts. I believe stress-buffering would be the correct term. Do not fear of being alone, mortal, weak, stupid or wrong -god is with you, has always been and always will be, whatever you do and wherever you go. How 's that for a reassuring thought? (What's that? You don't actually hear or see god, and feel no connection ? Try harder. There, see? This legitimate feeling of awe you have for nature, superimposed with the love you feel for your parents? Well now we will call this His Presence.)

There is a strong social component to both

Smokers of all nations have a very easy way to engage in a conversation (asking for a lighter, for a cigarette, an ashtrail), and a good reason to enjoy it -they share the same pleasure at the same time. This makes for a quite privilegiate form of contact which further tends to exclude non-smokers who might be disturbed by the smoke.

Nobody will be surprised to hear that the same holds for religion. Even in its less dangerous forms, religious people develop in communities. At the fundamendalist end, we get dangerously well-organised clubs.

They spread and work best on gullible people (young and/or non-educated and/or less intelligent)

There are well documented anti-correlations between religiosity and intelligence, and also between religious fervour and education. Moreover it is deplorably common to see young children and their much more malleable minds enrolled into more or less intrusive religious programs or teachings.

Again, it is no big news that the impact of cigarette on teenagers is just massive -and also not uncommon in younger persons. What I find interesting though, is that teens with high Emotional Quotient are less prone to smo king.

Both are damaging for one's health, and for other's.

The link between tobacco and all sorts of cancers, especially lung cancer, is established beyound doubt.

What is striking however is that religion and tobacco are both so addictive that even when aware of the harm they cause, many choose to pursue them to their very deaths. I don't think Marlboro's poor cowboy would have denied that:

And neither would the 9/11 terrorists

Pierrot

2009-11-14T22:49:00.005+01:00

Here's another beautiful french song. It's been out for a while, but for some reason I only stumbled upon the music video recently. It's from Loïc Lantoine, who I gather has morphed into a collective of four showmen (?!).

The song tells about friendship and how it can bring solace. In the song, Pierrot never fails to come in times of despair to offer the singer some kind of ineffable support, each time embodied in the delicate and pure guitar arpeggios that end the choruses. The clip actually makes this friend much more ambivalent and imaginary, playing with an old french lullaby.

Physics Haïku (2): the lattice

2009-10-27T00:01:00.001+01:00

"your wonder
joins mine,
on nature's lattice, one line"

Physics Haïku (1): cascade

2009-10-26T23:55:00.003+01:00

"proud water beads
laugh down two slopes
-foolish space, steep entropy"

SCIENCE vs RELIGION: how to answer the "many scientists are believers in god" argument?

2009-08-16T11:11:00.009+02:00

Implying that science and religion are compatible. I would suggest, for a short answer:

"Many physicians also smoke" (yes, even in your country)

Does it have to mean that smoking is good for health?
For a longer refutation, see Sam Harris's recent post. See also Atkins's site for nice quotes!

When plants are watching...

2009-04-23T11:08:00.009+02:00

... when birds manipulate humans, and ants can estimate Pi.

Writer Kevin Kelly has a very interesting post on intelligence in nature.
I would recommend to go there for the collection of amazing facts, and the quality of the writing, but I have some reserves on the general vision of intelligence I think he's putting forward.

Finir sa thèse ! (Finishing one's thesis)

2009-04-16T18:51:00.002+02:00

Beautiful lyrics, brilliant and fun (mmm..chords borrowed from Blu Cantrell it seems)

Le Minotaure

The author/singer/guitar player, Simon Berjeaut, is just ridiculously talented.
There is definitly some Brassens here, and a touch of Renaud but without the crude voice and lyrics.

Taken together with some recent discoveries heard on this radio, no, really, I think there's hope again for French music !

xkcd, graciously smart..

2009-04-13T10:32:00.007+02:00

..and "a treasured and carefully-guarded point in the space of four-character strings" according to its creator Randall Munroe.

Lexical semantics (3/4)

2009-04-12T17:20:00.010+02:00

How does it work?

We should be a little more careful in the way we phrase this question.
More specifically, what we want to describe in this post is: "How does the meaning of a word become available upon its (say, visual) presentation?".

There are a number of computational ways to tackle this question, and an important one is the PDP connectionist approach championned by researchers like Jay McClelland, Mark Seidenberg, David Plaut, David Rumelhart and Caralyn Patterson. All PDP models share the assumption that words are mappings between different codes; visual codes, sound codes and meaning codes. In neural network terms, this means that words are embodied in connections between different subsystems of units. To my knowledge, the most recent PDP model was proposed by Plaut and Boo th (2000) and it uses a fully recurrent network for the meaning subsystem: that is to say, a subnetwork in which all units are connected to each other. Recurrent networks are dynamical systems which exhibit attractor dynamics: the meaning of a word corresponds to an attractor of the system, i.e. a state towards which the system will naturally converge if it starts close enough. The central disk in the picture is an attractor in the state space of a system evolving under discrete time steps: each fan corresponds to many starting states being mapped to one, but only one point (the center of the disk) maps to itself -there's no escaping from it and it attracts all other states.

This contrasts with another connectionist variant supported by researchers like Jonathan Grainger, Max Coltheart and Colin Davis to cite a few. These also use neural network models, but words are now envisaged as units by themselves, not as connections between units. Access to a word's meaning is modelled as the activation of a dedicated meaning unit, exactly like access to a word's orthography activates a dedicated orthography unit. Although such systems have been described for instance in Carreiras et al (1997) or Coltheart et al (2001), to my knowledge an implementation has yet to be provided. The structure of such "localist" semantic networks has also been studied more recently, for instance in Steyvers and Tennenbaum (2005) (left picture), but this was done by abstracting away any network dynamics and thus any question about lexical access.

Yet another connectionist approach, more developmental and certainly more easily related to the brain, is the self-organising approach. Again we are talking about mappings between orthographic, semantic and phonological codes, but here codes of each type are supported by particular subsystems called self-organising maps (see right image), and these communicate with one another through hebbian connections. Examples of this approach include Mikkulainen (1993), Li et al (2004) and Mayor and Plunkett (2008). These models take a little bit from both worlds. Indeed, access to meaning corresponds to the apparition of a pattern of activation in the semantic map, an activation pattern which is distributed across all units like in the PDP approach, but which is centered on a so-called "best-matching" one like in the localist approach. These models usually put the emphasis on developmental and neural plausibility, and they are not yet ripe (or not upgraded) for the modeling of adult lexical decision or semantic categorisation tasks.

Finally we have semantic space models. These are not connectionist models: they don't need any neural network to be described. Moreover these models say nothing about the interaction between orthography, semantics and phonology, and hence they should not be seen as lexical access models, but rather as models of how humans build and represent lexical semantic information. The common hypothesis made by the tenants of this approach is that semantic codes are large vectors (they live in a high-dimensional semantic space), computed by more or less complicated co-occurence statistics in the language environment. Examples include Landauer and Dumais (1997)'s LSA model, and more recently Jones and Mewhort (2007)'s Beagle model (left picture). The focus of interest here lies in the internal organisation of the model: the topology of the semantic space -how related word meanings come to be represented in similar patches of the semantic space (the picture shows a 2D-projection of a semantic space in which related words are clustered together). Because they make no commitment as to how lexical access occurs, semantic spaces could theoretically be used with any of the previously mentioned models. However being distributed, semantic space vectors are more easily interfaced with the PDP or the self-organising approaches.

In summary we have a variety of ways to think of lexical semantic information, depending on exactly which findings we wish to account for. Do we seek to explain developmental or adult data? To what extent does neural plausiblity matters? Should we let facts about the brain interfere with our modelling of cognition? -not necessarily if one adheres to "multiple realizability"! What we can say however is that the connectionist approach, despite being a very versatile one, still constitutes the privileged way to describe lexical semantics. The importance of using high-dimensional spaces is also emerging. Also the literature is still highly polarized in the developmental vs mature dimension, with few information exchange taking place between models in each realm.

In another post, I'll try to explain how a less schizophrenic modelling appoach could exist, mediated by self-organising models.

Gigantic!

2009-04-09T13:37:00.003+02:00

..Well, that is, apart from the "insignificant bit of carbon" part at the end. I think this is a misconception. It reminds me of Stephen Hawking's "chemical scum" statement.. I agree with David Deutsch that this is the product of a limited vision of the human condition !

Change blindness

2009-04-03T10:07:00.007+02:00

A nice card trick!

Lexical semantics (2/4)

2009-03-24T11:06:00.007+01:00

Let us work our way up to the next question.

Is it located at any given place in the brain?

Well, at first the answer seemed to be yes. Wernicke's area has been recognized as early as in 1874 to be associated with word meaning. Patients with lesions in this area are afflicted with a kind of condition which leaves them unable to grasp the sense of words or to produce meaningfull speech, though importantly they are essentially unimpaired in other cognitive functions. Wernicke's area is close to the auditory cortex, but the associated deficit is not limited to the oral modality and extends to written language comprehension.

But of course and as often in cognitive science, things are a great deal more complicated. Researchers now agree that semantic retrieval involves at least a network of distant regions lying in the left frontal and temporal lobes, as suggested by the picture on the right (see this 2004 brain imagery study from Noppeney et al and this Nature neuroscience review from Patterson et al -both feature authoritative collaborators).

Characterizing the neural correlates of semantics remains work in progress, and as a matter of fact entire research labs are commited to this pursuit. Let me point to one important lab at the university of Wisconsin, which should also get the credit for the artwork.

Lexical semantics (1/4)

2009-03-23T10:25:00.008+01:00

What we cognitive scientists usually mean by lexical semantics is the system responsible for storing the meaning of words. How is this system organised? How does it develop ? How does it work? And is it located anywhere in the brain in particular? How can we know? Let's start with the last question.

How can we know?
Well cognitive science uses three main classes of tools to probe the human mind: behavioral experiments, brain imagery and computational models. There is of course a continual conversation going on between the three. Behavioral experiments ask each subject in a large group to operate a simple task in controled and reproducible laboratory conditions. Responses (usually motor responses) are then treated statistically and in this way significant effects can stand out. Hypotheses are suported or disproved.

A most common behavioral experiment in lexical semantics would be semantic categorization, in which upon presentation of a word on the screen, subjects are being asked whether it belongs to one category (is this bigger than a brick?) or not. What we look at: reaction times and percent errors. You can see that the idea is to plug our measuring apparatus to the simplest and hopefully the most objective signal possible.

Brain imagery simply tries to push the logic forward: we plug our measuring apparatus directly to the brain (non-invasively of course). The brain emits all kinds of signals strong enough to be picked-up at the surface of our skulls. There are many ways to do that: EEG, or MEG, another is PET, but I guess fMRI would be the most used brain imaging technique to date. All techniques have their strengths and weaknesses, for instance they are more or less accurate as for the spatial or temporal qualities of the signal.

Once behavioral and brain imagery studies are available, computational models try to make sense of the data. Computational models can be more biologically or psychologically oriented. A good balance in this respect seems to have been found early on by connectionism, which I for one define as the minimal concession to the hardware that one should acknowledge when modeling the software: using a large number of simple units connected to one another and which can be more or less active.

The explanatory power of a computational model is, as usual, given by the number of facts it can explain relative to the number of hypotheses it makes. But explanatory power is not the whole story, one would also like a model to make predictions which could be further tested with behavioral or brain studies -we want our models to be falsifiable.

Brilliant

2009-03-18T10:11:00.002+01:00

Watch evolution EXPLODE:

(via Asymptotia)

This just in

2009-03-14T16:00:00.003+01:00

Ouch. Mountazer Al-Zaïdi, the Iraqi journalist famous for his wonderful shoe throwing at Bush, has been sentenced to three years in jail.. (french article).

I hope he will eventually benefit from some kind of amnesty. The guy is huge in Irak and elsewhere, and presumably the pressure of international media could help. In fact we can help right now.

It takes a lot of courage to do what he did, and three years is quite a lot of time.

Links of the day

2009-03-13T10:55:00.004+01:00

Coachsurfing seems like a nice way to travel, since one is directly plugged into the life of local people, and the network has now reached 10^6 of them.

Although..we have friends in a similar network, globalfreeloader and they have not always been exstatic about it.

Perhaps there's an equivalent community but in which one doesn't have to share one's flat..? Well sure enough there is -ahhh, how I love the 21st century!

Working memory and IQ

2009-03-12T11:35:00.004+01:00

Nice article on the times today about brain training.

The authors quickly jump to working memory, and present recent experiments with subjects whose IQ scores increase with the amount of training they get in a dual n-back task. Another group of subjects, without training, did not improve their scores.

These results are interesting, but the authors correctly point out that the control condition is not appropriate to conclude in favor of the involvment of workng memory.

Sampling performance

2009-03-09T10:25:00.005+01:00

Kutiman has done some really spectacular sampling job, exclusively based on you tube videos.
In addition to his technical mastery, he seems to be just as confortable with funk, ragga, jungle, soul etc... It is a very generous work. As for me, I especially like sample 3, at 4:36. (via Cédric)

On a totally different subject, there is also a very nice post over at "When in doubt, do", where we learn that the afore mentioned Charles Darwin was a list maker, and mind you, for not so trivial matters!

DARWIN

2009-03-08T15:42:00.004+01:00

Via Asymptotia, via Wonga Blog.

Growing maps

2009-03-06T23:13:00.006+01:00

I'm currently looking a little bit closer at growing self-organising maps for a grant proposal. These are derived from the famous Kohonen map, a kind of neural network which establishes a correspondence between a high dimensional input space and a lower dimensional one (almost always 2 dimensions, but not necessarily). Whereas soms work with a fixed number of nodes, growing soms unfold with time. This makes them useful for e.g. language acquisition modeling (especially since these are unsupervised networks).

Wonderful memes

2009-03-05T09:28:00.008+01:00

We had a splendid dinner yesterday night at the Butte-aux-cailles with some family.

The occasion was my recent PhD defense, but we also discussed tons of other subjects, especially in relation to history and art. Knowledge flows from my grand-aunt and uncle like water from a source: crystalline, distilled and enriched by years of travels. I should add that both have very impressive carreers behind them, at the Louvre for my grand-aunt and as a teacher/civil servant at the ministry of Industry for my grand-uncle.

What may have triggered it all, this time, was my not so innocent trivia that Rome, Istanbul and Paris are said to have seven hills each. After a quick count we were only able to locate 6 hills in Paris, 2 in Istanbul and 5 in Rome -well it seems Paris really has only six.

Then all bets were off, and in no time we were discussing travels, hearing about unexpected adventures, being told about grandiose peaces of architecture and to my delight simply submerged under roman history. The successive emperors Trajan, Hadrian and Marcus-Aurelius were of particular interest, and we wondered whether or not Hadrian was right to let one piece of the empire go -soon enough we agreed he was.

Swimming

2009-03-03T17:33:00.005+01:00

There is something with swimming...which I simply don't experience with any other sports.

Working out at the Gym is pretty much about what the name indicates: work. No real fun involved -except perhaps sometimes when practiced in tandem. Running is not very fun either, but at least one can enjoy the landscape and after a while a kind of meditation can take place -like a discourse between mental and physical abilities- which is really great. But only by swimming do I entirely switch off from a day of work.

This is because suddenly and for an hour or so, I don't operate in the same element anymore. Sounds have this strange distortion caused by their propagation in the liquid medium, light is diffracted, water exerts more pressure on my skin, but my weight is reduced and my balance modified -these all conspire to reset the cognitive machinery. I always come out of swimming with an open mind and clear thoughts.

Ok, off to the swimming pool! By the way, I'd like to try an endless one some day.

Paris!

2009-03-02T23:00:00.005+01:00

This is "Le petit palais", close to the "Avenue des Champs-Elysées". Let us not forget how beautiful it is, even though we spoiled parisians never wander past it anymore.

Taken by my wife, more and more inspired with her new camera!

What's catastrophic forgetting?

2009-03-01T11:02:00.003+01:00

I should describe what catastrophic forgetting means in the context of cognitive science. Well, you'd think there would be a wikipedia entry for this but there's not -should probably go and write it down myself, meanwhile here's a nice introduction paper.

Anyway, catastrophic forgetting is the phenomenon by which some neural networks completely forget past memories when exposed to a set of new ones.

Naturally, this has been of some concerns for proponents of such systems which after all aim at simulating human memory functions. Humans do not, under normal circumstances, show this behavior. However it has been suggested and pretty convincingly argued that interleaved learning could circumvent the issue, at least in feed-forward networks trained with the backpropagation rule.

There is also a more catastrophic sense in which the notion has been used, and it is to describe the complete loss of any memories, past or recent, that occurs in e.g. the Hopfield network when exposed to undue numbers of patterns (interested in reading a recent thesis on the subject?). This is a subject I find utterly fascinating, having such ramifications as the function of dream sleep and palimpsest learning. More on that later.

Incipit

2009-02-28T22:09:00.003+01:00

Birth of this blog!
Catastrophic forgetting is intended as a window into the life of a cognitive scientist, featuring miscellaneous and more or less personnal thoughts, as well as regular reports on what's going on in computational cognitive modeling.