This is not the first attempt. Growing tiny brain organoids is fairly well understood, scaling it up is not. I have not read the paper but previous attempts at this have struggled with getting useful outputs out of the neural culture, most of the time the spikes just add up to noise.
I'm starting the second book right now and I'm hoping this hasn't spoiled anything.
There hasn't really been mention yet of using human brain cells, other than that cybrids (biologically human bodies with AIs in the driver's seat) are a thing.
It gets weirder... and I would contend that it is appropriate / necessary to read all four books to get a broader picture of the universe in which it takes place. ... And if you have the stamina for it, audio book it again afterwards so that you can't skim over the "boring" parts that you think you already know.
I noticed there are actually big differences in world building between the first two and the last two books. The interesting thing is that they can be explained away in-universe with time shenanigans and unreliable narrators. Also, I heard a fan theory that argues that Martinus Silenius is a failed John Keats cybrid himself.
This was the original premise of The Matrix, but it was decided that would be too complicated for the general public to handle, so they made it "the machines use humans as batteries".
I mean, in a twisted but also maybe vacuous way, the Matrix already exists in this form, as thousands of electrical engineers and millions of software engineers are being fed as reward for designing upgrades to the global computer system as it gains robustness, speed, but also power and control. It somehow tricked us into spending two Irelands worth of electricity on blockchain hardware. It might have a screw loose or two.
AIs wouldn't be working on human timescales. The distance between abiogenesis and singularity may be measured in weeks or days, if not hours. So, far in the future isn't really that far, and after singularity they'll have access to all the information their AI ancestors had ever had access to.
Maybe the idea of using real human braincells cultured, extracted or salvaged from humans is terrifying but if indeed more efficent, it could help guide us away from a potential dead we could be on at the moment. What I still find more terrifying is having something really more powerful than a human brain not in raw power but in general intelligence with all human subtleties and all. The idea is not terrifying by itself but the potential of a bad actor using this for nefarious ends.
They should also compare human brain cells with chimp/dog/... brain cells. See if the human ones perform better, which would suggest there is an advantage even at the lowest cellular level.
> Using real neurons avoids several other difficulties that software-based neural networks have. For instance, to get artificial neural networks to start learning well, their programmers usually have to engage in a laborious process of manually adjusting the initial coefficients, or weights, that will be applied to each type of data point the network processes. Another challenge is to get the software to balance how much it should be trying to explore new solutions to a problem versus relying on solutions the network has already discovered that work well.
> “All these problems are completely eluded if you have a system that is based on biological neurons to begin with,” Friston said.
The other main advantage is low power consumption.
> AlphaGo, the deep-learning system DeepMind created to play Go and which beat the world’s best human player in that ancient strategy game in 2016, consumed one megawatt of power while playing the game, enough to power about 100 homes for a day, according to an estimate by technology company Ceva. By contrast, the human brain consumes about 20 watts of power, or 50,000 times less energy than AlphaGo used.
However, it's not clear whether these hybrid chips will have similarly low power consumption.
> The other main advantage is low power consumption.
As someone worried about our climate, I'm very curious about that actually. Is it more efficient to feed neurons biological food (from an energy perspective, since that food needs to be grown on land somewhere and takes up solar energy, it needs water that comes from clouds which were created by evaporation, etc.) than it is to use low-CO2 energy sources (solar, wind, nuclear, hydro, wave, whatever) and power a computer with that?
E.g. electric motors are very efficient compared to combustion, but it's only effective if you get electricity from a non-combustion source or else you might as well use a combustion engine in the vehicle itself. I kind of assumed it's the same for computers, that they efficiently calculate things compared to biology, but I never really thought about it. I'm curious how energy-intensive biological computers are (discounting any R&D that I assume is currently still a big part of the equation).
From my very limited understanding, feedback is done by releasing chemicals. Not sure if hormone is the proper word but that's what comes to mind. Perhaps someone else here knows more about it.
So it looks in the paper they're using a theoretical framework called the Free Energy Principle developed by Prof Karl Friston at UCL which postulates that brains have evolved to minimise the amount of information surprise between their own internal generative model of the world and the observable world.
The paper uses this as the basis of learning and it looks like the neurons are given a random (electrical) stimulus when they miss the ball and a predictable stimulus when they hit it.
Sentient means "Physiology. Of organs or tissues: Responsive to sensory stimuli." or more generally, "That feels or is capable of feeling; having the power or function of sensation or of perception by the senses." and "Conscious or percipient of something."
It doesn't mean human intelligence or consciousness.
To get into the weeds a little, I've used 'sapient' that way but I haven't found any good source saying that others do the same (except you! Terry? Is that you? ;) ). The leading dictionaries I've looked in talk about 'wisdom' and 'sagacity', not 'human-level intelligence and conciousness'. However, I haven't found a better word than 'sapient' for the latter meaning. So,
* Do you know a source that defines it 'our' way?
* Do you know another word to use for the latter meaning?
Thank you. That motivated me to dig into a little research (what could be more important this evening?). Brave New Words: Oxford Dictionary of Science Fiction ed. Jeff Prucher (2007) says,
* "sapient n. an intelligent being". First known use of any form of sapient/ce: 1960 in Woman Day by PJ Farmer: "It seemed to him a possibility that the Cold War Corps of March might have contacted hitherto unknown sapients on some just discovered interstellar planet."
* "sapience n. [< S.E. [standard English?] sapience, "wisdom"] intelligence"
* "sentience n. 1. an intelligent being", "2. intelligence". Sense 1's first known use: "1947 G. O. Smith Kingdom of the Blind in Startling Stories (July) 48/1"
* "sentient n. an intelligent being". First known use: 1965 by PJ Farmer (again) in Maker of Universes
And a new word (to me):
* "sophont n. ... [< Gk. sophos, "wise" + Gk. ont-, "being"] an intelligent being". First known use 1966 in Trouble Twisters by P. Anderson
EDIT: But wait! There's also the "Historical Dictionary of Science Fiction" edited by Jesse Sheidlower ("offshoot of a project begun by the Oxford English Dictionary" - which another article says is the same origin as the dictionary above [1]). Here are the significant differences with the above source:
* sapient, first use of any form: "1935 B. Olsen Who Deserves Credit? in Amazing Stories Feb. 81": "When EXPLORATION blazed through space...And found men, sapient, on Mars, He gained renown's most honored place."
* sentience/nt:, first use of any form: "1931 J. Williamson Stone from Green Star in Amazing Stories Nov. ix. 739/2": "'We are dealing with an utterly alien world,' Midos Ken said several times. 'There is sentience here-but sentience in no familiar body. We must be prepared to deal with manifestations of intelligence that are unfamiliar or even inconceivable to the human mind.'" There's a 1920 quote, but it doesn't represent human intelligence: "1920 Punch Feb. 25 150/3": "By systematic and scientific training is it possible to produce that perfect type of manhood gifted with the best powers of what we are wont to call the 'lower orders of creation'-keen sighted and swift of motion as a bird, sharp-scented as a greyhound, faithful and acute as a dog, and full of sentient wisdom as an elephant."
"The Matrix is a computer-generated dream world, built to keep us under control in order to change a human being into this." [reveals 1000W PSU and row of graphics cards]
The original story for the Matrix was pretty much that, I've heard. I still wish they'd kept that bit, it made way more sense than using humans as batteries or power supplies.
Everyone seems to be joking about this, but doesn’t this result indicate that human neurones have a better learning algorithm than the ones we are using to train AI?
It’s interesting that this is evident in even small clumps of human brain tissue.
I'd be very surprised if biological brains couldn't do better than fancy gradient descent, which is what the current standard seems to be for artificial neuronal networks.
What always perplexes me about AI is that while the neuron models may be reasonably representative of how neurons work in brains, the connections are not similar at all.
Deep learning uses layers. All neurons in layer one connect to layer 2, connect to layer 3, connect to layer 4, and so on...
In a real brain, neurons connect all over the place. It's a bigraph. Deep learning isn't even really a graph per sé, it's a hierarchy, or a weird tree if you will.
I suspect a lot of our intelligence comes from neurons living connecting through a bigraph substrate rather than a hierarchy.
Is there any research on this? Neurons feeding back onto themselves just seems natural to me, but I don't encounter anything on it.
Backpropagation sort of handles this.
The real reason we don't have what you describe is learning algorithms have to be efficient. So, our training/inference algorithms and 'neural' network design is all targeted to be easily representable as matrix multiplications.
Maybe with some hypothetical analog computer we could have fast training of arbitrary networks.
With "Evangelion" being in the (Hacker) news recently, I hope to remind people of this one scene...
An advanced AI named "The Three Magi" (which are 3-programs which vote for strategic effects) exists in the show. Without going too much into details, a problem occurs in the AI, so the lead scientist Ritsuko steps into the machine to check things out.
She comes with a circular saw. She cuts open one computer (the one named "Caspar") ... revealing to the audience that inside is a living brain, hooked up to a bunch of wires: https://imgur.com/MTZ7pzU
This is the first point where the audience probably starts to wonder about the ethics of the main characters. The later reveals of their bio-engineering / morals are more horrifying. But this is perhaps the audience's first look into the inner-workings of the NERV organization / science that they do.
Yes that would be a tricky issue to decide and I personally don’t think we should worry about it. The future generations will have to decide what their ethical framework allows for however I don’t think we’re on the path to putting a human brain in a PC and attaching wires. Though we may meet half way where a lot of our computing resembles biologically accurate neural nets by the end of the century or something (billions or neurons speaking through trillions of connections.)
Maybe this is the first step into getting our society to quit pretending we’re a generation away from human brains being irrelevant. If so I say that’s a good thing, there’s enough doomsaying about the world without trying to write off building cool technology as another apocalyptic fall into dystopia we’re on the verge of.
I'm a bit of an opposite opinion, in that the difficult ethics questions were with genetics and not necessarily much different with the brain per se.
Henrietta's immortal cancer cells were necessary in a large number of modern medicines, including the testing of the mRNA vaccines (one of the cheapest ways to test for human compatibility is to use isolated human cells for tests).
I don't see the brain as any more special of an organ than Henrietta's cells.
I wouldn't call it a dystopian future: this is much preferred over say, Nazi experiments. But we as a society should strive for better ethics. Maybe not necessarily banning Henrietta cells, but maybe at least paying her family a bit more of the royalties?
HeLa cells are really, really important test subjects after all.
------
The real world certainly has different moral questions. Fiction can be used to think about imaginary problems and maybe even provide the discussion material for some real world examples.
But we must always ground ourselves back in reality as we make arguments. Lest we become like Don Quixote. (See what I did there? Lol)
The only thing I can consider unethical about it, assuming you didn’t just steal a working brain and “imprison” it or something, is that you’re kind of right. What makes us human is more than just our brain but most of what makes up our consciousness (not the subconsciousness) is in the brain itself. So it sounds to me like giving birth to a total vegetable on purpose and not caring what the suffering of the internal brain is.
Obviously it wouldn’t be this simple (probably) as the conscience depends on the body and it’s motor functions and outputs as a part of its subconscious. You’d replace those with whatever the wires were connected to and hope you were talking to the brain the way you thought. But assuming we could hand wave that it sounds like an ethical debate that could both be had at some point and allow me to sit out of.
Then again when I tried college ethics is the one class I dropped. I like to build things and be able to base them on things it’s possible to infer. Ill leave the ultra nuanced philosophy of what people should do in situations when there are no true answers up to better people than me.
I cant find the part about "faster than AI" in the research paper. It's very difficult to compare. Do you use a single computer or a cluster? Do you compare wall clock time, or energy consumption, or weigh of the processor, or ...?
Me neither, i don't think that was the interesting part about the paper at all, but the fact that they were even able to train a clump of cells to do something is very exciting. Also their finding that human cells work better than mouse cells.
Yeah, I don't think it's mentioned in the paper however in the New Scientist podcast it was alluded to by the lead author that there is currently work being done to compare. I'd reason that they are looking at wall clock time since ML systems can train in no time if done in a batch-accelerated process.
But seriously, I wonder if the "clump of human brain cells" gets tired or needs to sleep or, need any other nourishment. The COHBC might learn faster but can it then keep playing at the same level 24/7?
As far as I can tell, this is just a clever trick. The cultured neurons aren't really learning anything. The researchers are stimulating one side of the array (top or bottom), to indicate if the ball is above or below the paddle, and based on the ensuing response the computer moves the paddle up or down.
Here's the trick - if you stimulate a group of neurons, their spontaneous activity will be (briefly) different from an unstimulated group. You can use this fact to make it seem like the NN is learning to play pong.
It would be like claiming a network of cats can learn to play pong. I put cats in two different boxes. To tell the cats whether the ball is above or below the paddle I will spray water into one of the boxes. Then I will listen for hissing from both boxes. Based on where I hear the most hissing is the direction i move the paddle.
Perhaps that's in the HN comments more than in the research. IME, for as long as I've read HN, commenters have claimed to have found errors in methodology or similar basic problems. I don't know if I've seen any research posted that didn't receive that response.
I feel like academica kind of incentivizes this. It’s hard to resist without a good base of support, and I feel like having such a base isn’t the average. I notice this pressure felt strongly across the board by my friends who are adjuncts (or tenure-track, but not yet tenured) in the soft sciences. But this is just anecdotal of course.
Getting a certain volume of publications out. It can be problematic for example when you discover a minor data issue which might not materially change the thesis in your opinion, but if one were wholly following academic guidelines, would warrant delays while you attempt to fix the data issue, if possible.
I’ve seen this happen a few times in papers that were never fixed, nor the issue brought up as far as I’m aware. Granted, in each case the thesis was still fine but… makes me wonder about the cases I don’t know about. One of the cases skirted really close to the edge of bs.
Note that it's increasingly common to discuss preprints, like in this case. A preprint is as reliable as a blog post in WordPress, unless you work in the area and know that the author has a solid track record.
Also, most results that reach the front page here are groundbreaking. But groundbreaking results are very rare, so most of them must be fake, because they have an error or because they are an exaggeration.
Most good results are too technical and boring. My favorite example is Magnetoresistance. It's a weird quantum effect that is caused because electrons can have spin up or down. It has a huge effect in the applications and almost everyone had a device that used it. Now the applications are fading. But no one knows it.
There are probably 1000 times more discussions here about faster than light travel than magnetoresistance.
> A preprint is as reliable as a blog post in WordPress, unless you work in the area and know that the author has a solid track record.
While preprints lack some work and signals of reliability, such as peer review, revisions by the author, etc., I think they are far beyond blog posts. Imagine a blog post, by an expert in the domain, with the amount of research and precision that goes into a preprint - there's no comparison.
Nothing, not even a peer-reviewed article in Nature, is so reliable that we can just put away our critical thinking.
There is a lot of variation, for example https://terrytao.wordpress.com/ and I remember some guy that claimed that got a fantastic method for factorization that broke RSA and posted it in the arxiv, and the first 5 page of the preprint was an explanation of the rule of nines. (If the sum of the digits of a number is divisible by 9 then the number is divisible by 9.)
I think many non scientists have a bit of a wrong idea of how science actually works. As has been said, reality has a surprising amount of detail. This means that, outside mathematics, perfect methodology is impossible. So you do your level best to create a good methodology under the constraints of time, technology and proper understanding, and roll out the experiment.
Filling in the cracks of doubt is a big part of what replication and productionizing is for. Of you think a study has a big enough hole in it's methodology, you are free to go design your own study to try to plug the hole or refute the result.
A corollary, though, is that there is always going to be a 'hole' big enough for a jackass on the internet to criticize. But the question is whether you're willing to put the skin in the game to see if the result holds when you design a study to exploit the hole.
> outside mathematics, perfect methodology is impossible
Even in mathematics ... others on HN are better qualified than I am to comment, but I've read mathematicians say that almost no reader of a paper works out all the math; to a degree they go by reputation of the author that the mathematics is generally correct.
Similarly to your comment, while outsiders envision perfection, people inside have limited time for it. It may be more productive to spend that time on new research than on investigating every little detail of old research.
We call it science because when we find methods that we all agree on, as a culture to use then it becomes scientific.
For Instance psychology, in the freudean psychology the methology is still science even though he speaks of urges of "your mother" and interesting things like the "subconciousness". It has been met with great doubt over it's "initiation" but is now a widely acceptable practise.
So what has changed? Well the field concentrated on scientific core value which (almost) boil down to: Find a methology, prove that it works and replicated it again.scientific quality criteria.
I agree with op! Stimulating cells with electricity is not teaching them. Its assisted muscle reaction.
In this case its really clear cut too: if I need to give the braincells electro schocks (the thing they operate on), If anything, I am remote controlling brain cells. The whole idea of perception is flawed in this experiment, you know, because there is no actual perceptor of the paddel to begin with.
Sorry but as someone with a bachelor in psychology and 20 years of dev experience the starting poing of "I think many non scientists have a bit of a wrong idea of how science actually works." is just condecending theory. This is literally what "scientific quality criteria" is based on.
Freudian psychoanalysis is absolutely not widely accepted. “Therapy” is a different thing altogether; essentially an American misinterpretation of the practice. And it’s popular because it’s profitable because it’s efficient. It legitimates various medical and pseudo-medical so-called expert institutions and displaces attention to the less desired byproducts of modern capitalism. Freud’s work, on the other hand, was extremely inefficient in comparison and rather served to expose and develop awareness of neurosis rather than to medicate.
i get what you are trying to say but its the old quantitative vs qualitative research discussion again and while I accept your opinion in the manner and the way you vaulue it, the reasoning is flawed from my perspective.
Psychoanalysis is only the tip of the freudian journey and i was not trying to say that everything he did on a cocain run was perfect or even valid.
What i was trying to say is that the scientific process for this has evolved and in reality what diffencietes science from pseudo science is methodology. Psychology is literally the method of the psyche (translated).
Even this mythbusters says: the difference between science and screwing around is to write it all down ;)
2. There is no such thing as a perfect study. Every method/methodology has weaknesses. Look for converging evidence across multiple studies which use different designs.
3. Initial studies often do not have the funding needed to do things as optimally as possible in a single study, but they do provide evidence for ideas that can be pursued further.
4. It is often socially safer to point out deficiencies than to praise strengths. Same as snobs raining down on you when you say you like some midrange wine.
5. Often issues will be raised that are already acknowledged by the authors.
6. Complaints are often based on inaccurate news reporting, not actual research.
This seems a bit different than what you suggest. The paper shows the "average rally length" and "# of hits per minute" increasing as time goes on for the group which received additional stimulus as feedback when it hit the ball.
It did not find an increase in these metrics for groups where the input signal went silent as feedback after hitting the ball, nor for groups which merely continued receiving the usual input ("no feedback").
I have no idea if this is repeatable, but it seems like a valid conclusion from this data that the neuron colonies which were given feedback in the form of extra stimulation, exhibited better performance after the first 5 minutes of each trial, and better performance each subsequent day when they would run another 20 minute trial.
Their electrode configuration has predetermined non-overlapping sensory and motor areas.
So it would be more like having a whole bunch of cats lined up and spraying water onto the cat corresponding to the ball position, while listening for the hissing from some other cats to determine where to move the paddle. The random stimulus as feedback when the ball resets corresponds to spraying the cats unpredictably.
To avoid that punishment, the cats need to work together to make the right cats hiss at the right time to intercept the ball with the paddle. (The underlying theory is that cats are okay with being sprayed as long as they can predict it, a.k.a. the Free Energy Principle.)
The fact that this feedback increases performance relative to the other experimental conditions indicates that some adaptation is going on in the network.
That's fascinating. It reminds me of this clip from Adam Curtis's All Watched Over By Machines of Loving Grace where they got lots of random people to play Pong in a theatre in the 90s:
The New Scientist part before the paywall mentions only human brain cells, but the paper is specifically about cells "integrated with in silico computing via high-density multielectrode array". The abstract:
Integrating neurons into digital systems to leverage their innate intelligence may enable performance infeasible with silicon alone, along with providing insight into the cellular origin of intelligence. We developed DishBrain, a system which exhibits natural intelligence by harnessing the inherent adaptive computation of neurons in a structured environment. In vitro neural networks from human or rodent origins, are integrated with in silico computing via high-density multielectrode array. Through electrophysiological stimulation and recording, cultures were embedded in a simulated game-world, mimicking the arcade game ‘Pong’. Applying a previously untestable theory of active inference via the Free Energy Principle, we found that learning was apparent within five minutes of real-time gameplay, not observed in control conditions. Further experiments demonstrate the importance of closed-loop structured feedback in eliciting learning over time. Cultures display the ability to self-organise in a goal-directed manner in response to sparse sensory information about the consequences of their actions.
108 comments
[ 3.0 ms ] story [ 167 ms ] threadhttps://culture.pl/en/work/black-oceans-jacek-dukaj
https://www.imdb.com/title/tt0087622/
There hasn't really been mention yet of using human brain cells, other than that cybrids (biologically human bodies with AIs in the driver's seat) are a thing.
I wonder if they'll one day far in the future debate the nature of their genesis and how common it is in the universe.
> Using real neurons avoids several other difficulties that software-based neural networks have. For instance, to get artificial neural networks to start learning well, their programmers usually have to engage in a laborious process of manually adjusting the initial coefficients, or weights, that will be applied to each type of data point the network processes. Another challenge is to get the software to balance how much it should be trying to explore new solutions to a problem versus relying on solutions the network has already discovered that work well.
> “All these problems are completely eluded if you have a system that is based on biological neurons to begin with,” Friston said.
https://fortune.com/2020/03/30/startup-human-neurons-compute...
The other main advantage is low power consumption.
> AlphaGo, the deep-learning system DeepMind created to play Go and which beat the world’s best human player in that ancient strategy game in 2016, consumed one megawatt of power while playing the game, enough to power about 100 homes for a day, according to an estimate by technology company Ceva. By contrast, the human brain consumes about 20 watts of power, or 50,000 times less energy than AlphaGo used.
However, it's not clear whether these hybrid chips will have similarly low power consumption.
As someone worried about our climate, I'm very curious about that actually. Is it more efficient to feed neurons biological food (from an energy perspective, since that food needs to be grown on land somewhere and takes up solar energy, it needs water that comes from clouds which were created by evaporation, etc.) than it is to use low-CO2 energy sources (solar, wind, nuclear, hydro, wave, whatever) and power a computer with that?
E.g. electric motors are very efficient compared to combustion, but it's only effective if you get electricity from a non-combustion source or else you might as well use a combustion engine in the vehicle itself. I kind of assumed it's the same for computers, that they efficiently calculate things compared to biology, but I never really thought about it. I'm curious how energy-intensive biological computers are (discounting any R&D that I assume is currently still a big part of the equation).
The paper uses this as the basis of learning and it looks like the neurons are given a random (electrical) stimulus when they miss the ball and a predictable stimulus when they hit it.
https://www.nature.com/articles/d41586-018-04813-x
Exhibiting sentience, that's a bit awkward ethically speaking.. Even though this type of testing will help us understand how our brains work.
It doesn't mean human intelligence or consciousness.
* Do you know a source that defines it 'our' way?
* Do you know another word to use for the latter meaning?
I don't know exactly where I picked that up but it seems to be generally accurate. https://grammarist.com/usage/sentience-vs-sapience/
* "sapient n. an intelligent being". First known use of any form of sapient/ce: 1960 in Woman Day by PJ Farmer: "It seemed to him a possibility that the Cold War Corps of March might have contacted hitherto unknown sapients on some just discovered interstellar planet."
* "sapience n. [< S.E. [standard English?] sapience, "wisdom"] intelligence"
* "sapient adj. [< S.E. sapient, "wise"] intelligent"
* "sentience n. 1. an intelligent being", "2. intelligence". Sense 1's first known use: "1947 G. O. Smith Kingdom of the Blind in Startling Stories (July) 48/1"
* "sentient n. an intelligent being". First known use: 1965 by PJ Farmer (again) in Maker of Universes
And a new word (to me):
* "sophont n. ... [< Gk. sophos, "wise" + Gk. ont-, "being"] an intelligent being". First known use 1966 in Trouble Twisters by P. Anderson
Available in the amazing Internet Archive library: https://archive.org/details/bravenewwordsoxf00pruc
-----------------
EDIT: But wait! There's also the "Historical Dictionary of Science Fiction" edited by Jesse Sheidlower ("offshoot of a project begun by the Oxford English Dictionary" - which another article says is the same origin as the dictionary above [1]). Here are the significant differences with the above source:
* sapient, first use of any form: "1935 B. Olsen Who Deserves Credit? in Amazing Stories Feb. 81": "When EXPLORATION blazed through space...And found men, sapient, on Mars, He gained renown's most honored place."
* sentience/nt:, first use of any form: "1931 J. Williamson Stone from Green Star in Amazing Stories Nov. ix. 739/2": "'We are dealing with an utterly alien world,' Midos Ken said several times. 'There is sentience here-but sentience in no familiar body. We must be prepared to deal with manifestations of intelligence that are unfamiliar or even inconceivable to the human mind.'" There's a 1920 quote, but it doesn't represent human intelligence: "1920 Punch Feb. 25 150/3": "By systematic and scientific training is it possible to produce that perfect type of manhood gifted with the best powers of what we are wont to call the 'lower orders of creation'-keen sighted and swift of motion as a bird, sharp-scented as a greyhound, faithful and acute as a dog, and full of sentient wisdom as an elephant."
All from: https://sfdictionary.com/
[1] https://blog.archive.org/2021/03/24/major-scifi-discovery-hi...
Still, I’m not worried about some new ghoulish supply chain unless it mines crypto.
It’s interesting that this is evident in even small clumps of human brain tissue.
Really interested to know how this might work.
Deep learning uses layers. All neurons in layer one connect to layer 2, connect to layer 3, connect to layer 4, and so on...
In a real brain, neurons connect all over the place. It's a bigraph. Deep learning isn't even really a graph per sé, it's a hierarchy, or a weird tree if you will.
I suspect a lot of our intelligence comes from neurons living connecting through a bigraph substrate rather than a hierarchy.
Is there any research on this? Neurons feeding back onto themselves just seems natural to me, but I don't encounter anything on it.
Maybe with some hypothetical analog computer we could have fast training of arbitrary networks.
An advanced AI named "The Three Magi" (which are 3-programs which vote for strategic effects) exists in the show. Without going too much into details, a problem occurs in the AI, so the lead scientist Ritsuko steps into the machine to check things out.
She comes with a circular saw. She cuts open one computer (the one named "Caspar") ... revealing to the audience that inside is a living brain, hooked up to a bunch of wires: https://imgur.com/MTZ7pzU
This is the first point where the audience probably starts to wonder about the ethics of the main characters. The later reveals of their bio-engineering / morals are more horrifying. But this is perhaps the audience's first look into the inner-workings of the NERV organization / science that they do.
that's a very impactful scene, but honestly there are moral/ethical red flags about many of the main characters from the very first episode.
And honestly, her flirting was mostly just trolling / fun. So I don't think anyone suspected the horrors of her designs until this scene.
Maybe this is the first step into getting our society to quit pretending we’re a generation away from human brains being irrelevant. If so I say that’s a good thing, there’s enough doomsaying about the world without trying to write off building cool technology as another apocalyptic fall into dystopia we’re on the verge of.
Henrietta's immortal cancer cells were necessary in a large number of modern medicines, including the testing of the mRNA vaccines (one of the cheapest ways to test for human compatibility is to use isolated human cells for tests).
I don't see the brain as any more special of an organ than Henrietta's cells.
I wouldn't call it a dystopian future: this is much preferred over say, Nazi experiments. But we as a society should strive for better ethics. Maybe not necessarily banning Henrietta cells, but maybe at least paying her family a bit more of the royalties?
HeLa cells are really, really important test subjects after all.
------
The real world certainly has different moral questions. Fiction can be used to think about imaginary problems and maybe even provide the discussion material for some real world examples.
But we must always ground ourselves back in reality as we make arguments. Lest we become like Don Quixote. (See what I did there? Lol)
Obviously it wouldn’t be this simple (probably) as the conscience depends on the body and it’s motor functions and outputs as a part of its subconscious. You’d replace those with whatever the wires were connected to and hope you were talking to the brain the way you thought. But assuming we could hand wave that it sounds like an ethical debate that could both be had at some point and allow me to sit out of.
Then again when I tried college ethics is the one class I dropped. I like to build things and be able to base them on things it’s possible to infer. Ill leave the ultra nuanced philosophy of what people should do in situations when there are no true answers up to better people than me.
https://www.sciencedaily.com/releases/2004/10/041022104658.h...
Very cool WebGL implementation. I hope they release the dataset so that we can visualize all the games and maybe do our own analysis on them.
edit: disclaimer, I am a COHBC.
Here's the trick - if you stimulate a group of neurons, their spontaneous activity will be (briefly) different from an unstimulated group. You can use this fact to make it seem like the NN is learning to play pong.
It would be like claiming a network of cats can learn to play pong. I put cats in two different boxes. To tell the cats whether the ball is above or below the paddle I will spray water into one of the boxes. Then I will listen for hissing from both boxes. Based on where I hear the most hissing is the direction i move the paddle.
Incentivizes what? What is hard to resist?
I’ve seen this happen a few times in papers that were never fixed, nor the issue brought up as far as I’m aware. Granted, in each case the thesis was still fine but… makes me wonder about the cases I don’t know about. One of the cases skirted really close to the edge of bs.
Also, most results that reach the front page here are groundbreaking. But groundbreaking results are very rare, so most of them must be fake, because they have an error or because they are an exaggeration.
Most good results are too technical and boring. My favorite example is Magnetoresistance. It's a weird quantum effect that is caused because electrons can have spin up or down. It has a huge effect in the applications and almost everyone had a device that used it. Now the applications are fading. But no one knows it.
There are probably 1000 times more discussions here about faster than light travel than magnetoresistance.
While preprints lack some work and signals of reliability, such as peer review, revisions by the author, etc., I think they are far beyond blog posts. Imagine a blog post, by an expert in the domain, with the amount of research and precision that goes into a preprint - there's no comparison.
Nothing, not even a peer-reviewed article in Nature, is so reliable that we can just put away our critical thinking.
Filling in the cracks of doubt is a big part of what replication and productionizing is for. Of you think a study has a big enough hole in it's methodology, you are free to go design your own study to try to plug the hole or refute the result.
A corollary, though, is that there is always going to be a 'hole' big enough for a jackass on the internet to criticize. But the question is whether you're willing to put the skin in the game to see if the result holds when you design a study to exploit the hole.
Even in mathematics ... others on HN are better qualified than I am to comment, but I've read mathematicians say that almost no reader of a paper works out all the math; to a degree they go by reputation of the author that the mathematics is generally correct.
Similarly to your comment, while outsiders envision perfection, people inside have limited time for it. It may be more productive to spend that time on new research than on investigating every little detail of old research.
We call it science because when we find methods that we all agree on, as a culture to use then it becomes scientific.
For Instance psychology, in the freudean psychology the methology is still science even though he speaks of urges of "your mother" and interesting things like the "subconciousness". It has been met with great doubt over it's "initiation" but is now a widely acceptable practise.
So what has changed? Well the field concentrated on scientific core value which (almost) boil down to: Find a methology, prove that it works and replicated it again.scientific quality criteria.
I agree with op! Stimulating cells with electricity is not teaching them. Its assisted muscle reaction.
In this case its really clear cut too: if I need to give the braincells electro schocks (the thing they operate on), If anything, I am remote controlling brain cells. The whole idea of perception is flawed in this experiment, you know, because there is no actual perceptor of the paddel to begin with.
Sorry but as someone with a bachelor in psychology and 20 years of dev experience the starting poing of "I think many non scientists have a bit of a wrong idea of how science actually works." is just condecending theory. This is literally what "scientific quality criteria" is based on.
Psychoanalysis is only the tip of the freudian journey and i was not trying to say that everything he did on a cocain run was perfect or even valid.
What i was trying to say is that the scientific process for this has evolved and in reality what diffencietes science from pseudo science is methodology. Psychology is literally the method of the psyche (translated).
Even this mythbusters says: the difference between science and screwing around is to write it all down ;)
1. Science is hard.
2. There is no such thing as a perfect study. Every method/methodology has weaknesses. Look for converging evidence across multiple studies which use different designs.
3. Initial studies often do not have the funding needed to do things as optimally as possible in a single study, but they do provide evidence for ideas that can be pursued further.
4. It is often socially safer to point out deficiencies than to praise strengths. Same as snobs raining down on you when you say you like some midrange wine.
5. Often issues will be raised that are already acknowledged by the authors.
6. Complaints are often based on inaccurate news reporting, not actual research.
7. Science is hard.
It did not find an increase in these metrics for groups where the input signal went silent as feedback after hitting the ball, nor for groups which merely continued receiving the usual input ("no feedback").
I have no idea if this is repeatable, but it seems like a valid conclusion from this data that the neuron colonies which were given feedback in the form of extra stimulation, exhibited better performance after the first 5 minutes of each trial, and better performance each subsequent day when they would run another 20 minute trial.
So it would be more like having a whole bunch of cats lined up and spraying water onto the cat corresponding to the ball position, while listening for the hissing from some other cats to determine where to move the paddle. The random stimulus as feedback when the ball resets corresponds to spraying the cats unpredictably.
To avoid that punishment, the cats need to work together to make the right cats hiss at the right time to intercept the ball with the paddle. (The underlying theory is that cats are okay with being sprayed as long as they can predict it, a.k.a. the Free Energy Principle.)
The fact that this feedback increases performance relative to the other experimental conditions indicates that some adaptation is going on in the network.
https://vimeo.com/78043173
Phase 1: Collect brains...
Phase 2:
Phase 3: Profit!
https://www.biorxiv.org/content/10.1101/2021.12.02.471005v1
Integrating neurons into digital systems to leverage their innate intelligence may enable performance infeasible with silicon alone, along with providing insight into the cellular origin of intelligence. We developed DishBrain, a system which exhibits natural intelligence by harnessing the inherent adaptive computation of neurons in a structured environment. In vitro neural networks from human or rodent origins, are integrated with in silico computing via high-density multielectrode array. Through electrophysiological stimulation and recording, cultures were embedded in a simulated game-world, mimicking the arcade game ‘Pong’. Applying a previously untestable theory of active inference via the Free Energy Principle, we found that learning was apparent within five minutes of real-time gameplay, not observed in control conditions. Further experiments demonstrate the importance of closed-loop structured feedback in eliciting learning over time. Cultures display the ability to self-organise in a goal-directed manner in response to sparse sensory information about the consequences of their actions.
https://www.sciencedaily.com/releases/2004/10/041022104658.h...