Amateur neuroscientist and curious cat
Homepage: https://coldcoffee.neocities.org/
dmac_93
Karma: 65
There is an economy of attention, and the AI does not understand how I want to spend my very limited free trial of your attention.
The AI puts many big words where a simple one would do. It rambles and it encourages authors to ramble.
I agree with the conclusion that LLMs arent conscious, but Seth Anil’s thinking seems muddled, and the video description has a suspicious number of em dashes.
The first reason, self-projection, is not actually evidence against the theory that LLMs are conscious.
The second reason is phrased as though he thinks a simulation of a human brain wouldnt be conscious, even if it included all of the electric and chemical properties of the brain.
As for the rest, well, ive nothing nice to say. I dont think he understands what hes talking about.
Ostensibly, if only the police are well fed then they’ll go to work and enforce the rules. But I worry that a critical mass of people will get hungry and angry. My worry is that when the brain gets too hungry it turns off the smart parts and reverts to its insincts. The cortex take a lot of energy to run at full speed. The brain is setup so that if the cortext fails -for whatever reason- then the insincts keep working on their own. Even if hangry ppl dont burn it all down in the first two weeks, I think hangry ppl suck to be around in ways we are not accustomed to?
If i could pull a nugget of truth out of SMST’s work, it would be that the brain is a control system. There are many different types of control system and the brain probably uses all of them. For example the spinal cord alone contains closed loop controllers (for controlling muscle forces and positions), open loop controllers (for pain withdrawl reflexes), and finite state machines (for walking & running on four legs).
The question is, how does the brain use RL to implement a control system? And how does that interface with the other control systems in the brain?
My theory is that the brain uses both reinforcement learning and closed loop control. Then the brain uses the closed loop controller’s error to generate reward signals endogenously.
That is to say: a reward is given when the closed loop controller reaches its setpoint, and a penalty is given if it moves too far from its setpoint.
The seats in minivans fold down into the floor these days, which IMO justifies having the extra height. Its a very convenient and general purpose feature. But being high does add weight and construction cost and rollover risk.
No love for New England? We’ve got good education and a solid base of tech companies.
An alternative narrative is that AGI is the product of a long slow slog of research into computational neuroscience. In this scenario the ambitious ppl in NYC and Cali refuse to research the right stuff because the rewards are too far off. And here is where I see New England as having good vibes, for working on basic research that has no immediate applications.
This is a sermon I wrote.
Hello. Today I’d like to talk about something a bit different. I’d like to talk about theology, my own personal creation myth: something to explain why the world exists, when it seems like the world could just as easily not exists. I posit that the universe is a mathematical construct, and that all mathematical constructs inherently exist.
Now, First I’d like to a take a moment to define what exactly math is and to appreciate some of it’s properties. Broadly speaking, mathematics is the study of rules and rule-based systems. The basic premise is that you start with some very simple rules (called the axioms) and working from these rules you can prove other more complex facts about the system.
And what’s important is that all of the rules are being followed all of the time. At every step along the way, and at every point in the system, the rules are uniformly and consistently applied.
If ever any of the rules are broken, then the whole system becomes meaningless. For example if you can prove that “0 == 1” then, with some trivial rearrangements of the equations, you can show that every number is equal to every other number, and also that no number is equal to anything else, not even itself. The introduction of an invalid equation into a mathematical system results in an infinite number of non-sensical results.
Clearly, you can not have a meaningful existence with broken or inconsistent mathematics. Contemplate with me, an existence with broken equations. You would be unable to count the things that exist, and even the most basic of facts about our reality would become unhinged.
But that is not the world we live in. In our universe math is eternal and omnipresent. An equation like “1 + 1 == 2” is as true today as it was true yesterday and it will be true tomorrow. There is no corner of space or point in time where this equation is wrong.
Now I’d like to mention another property of mathematical systems; that once you have a system of equations, you can rearrange and combine those equations in order to introduce new equations to the system. Assuming that you did the math correctly, then the new equations are going to be valid members of that system of equations, and this is a direct consequence of the rules of mathematics.
In a way, those new equations were always there, even if you were unaware of them. When you prove a new fact about the world, that fact was already true even before you proved it.
When we think about systems of equations, we like to think about small finite sets of equations, but really every system of equations contains infinitely many equations, and what we’re doing when we do arithmatic is we’re searching for one equation in specific, out of all of those infinite equations.
So what we think of as a system of equations is actually just some arbitrary a subset of the set of all valid equations. So if you have any mathematical system, then in fact you’re working with all mathematical systems, in a way.
Now let’s discuss our world, and where it came from. I believe that our physical universe can be perfectly modeled using mathematics, that there exist some equations which accurately describe everything in our universe, down to the smallest iota of matter and all of the forces acting upon it.
And I’m not talking about Newton’s or Einstein’s laws or any such crude approximations of our physical reality, but rather a perfect description of the universe which yields 100% accuracy.
If we assume that such equations do exist then they are part of the set of all valid equations. And if on the other hand, our universe does not follow any logic or abide by any rules then, as we already saw, it would be very difficult for anything to meaningfully exist in our universe.
Now back to the question of creation, the question of “why does anything exist” can be reduced to the question of “why does mathematics exist?” And does mathematics really need to justify its own existence? Math is simply the idea that you can have rules, and that what follows is a consequence of following those rules. And you don’t even need to follow the rules, it’s just that if you don’t follow the rules you get meaningless results.
Mathematics is just an abstract concept. I do not think that mathematics needs to justify its own existence, given that it does not have any tangible or concrete implementation. And yet, starting from this abstract concept of mathematics, we can prove that our very real universe exists as a direct consequence. Our universe is a set of physics equations embedded inside of the set of all valid equations, and with us animals embedded inside of the physics of our universe.
Our world exists because the equations that describe our world exist. From one point of view, it may seem like I’m equating all of existence with an ethereal abstract concept which does not really concretely exist, but from another point of view I’m saying that we must exist as an inevietable result of the very concept of rules and rule-based systems.
In conclusion, this creation myth places very few constraints on the physics of our universe; it allows for worlds that are deterministic or probabilistic; discrete or real valued; computable or containing the number PI; finite or infinite! It even permits conventional religions to exist, assuming of course that the gods are mathematically consistent. And finally all of this implies that other universes exist, goverened by different sets of physics equations, and, undoubtedly, some of those other universes also support intelligent life.
I’d like to share a book recommendation:
“Writing for the reader”
by O’Rourke, 1976
https://archive.org/details/bitsavers_decBooksOReader1976_3930161This primer on technical writing was published by Digital Equipment Corporation (DEC) in 1976. At the time, they faced the challenge of explaining how to use a computer to people who had never used a computer before. All of the examples are from DEC manuals that customers failed to understand. I found the entire book delightful, insightful, and mercifly brief. The book starts with a joke, which I’ve copied below:
On the West Coast they tell the story of a plumber who started using hydrochloric acid on clogged pipes. Though he was pleased with the results, he wondered if he could be doing something wrong. So he wrote to Washington to get expert advice on the matter. In six weeks he received the following reply:
“The efficacy of hydrochloric acid in the subject situation is incontrovertible, but its corrosiveness is incompatible with the integrity of metallic substances.”
The plumber, who was short on formal education but long on hope, was elated. He shot a thank-you letter back to Washington. He told them he would lose no time in informing other plumbers about his discovery. Five weeks later he got another message:
“In no case can we be presumed responsible for the generation of pernicious residues from hydrochloric acid, and we strongiy recommend, therefore, than an alternative method be utilized.”
The plumber was delighted. He sent his third letter in the next mail. In it, he said that about 15 plumbers in his city were now using hydrochloric acid for pipes. All of them liked it. Now he wondered whether the good people in Washington could help him spread the news of his discovery to plumbers throughout the country. At this point, the correspondence fell into the hands of a rare Washington bureaucrat—one who knew how to write to plumbers. Within a week the plumber was reading these words:
“Stop using hydrochloric acid. And tell your friends to stop too. It eats the hell out of pipes.”
Certainly the letters in this interchange are a far cry from technical writing. Nevertheless, they offer a lesson to the new technical writer: Write so your reader can understand.
Sutton seems to confuse intelligence with life. These are distinctly different concepts. Compare LLMs and bacteria: LLMs are intelligent but not alive, bacteria are alive but not intelligent. Bacteria have goals, such as consuming food and avoiding hazards, and bacteria take directed action to accomplish their goals.
Banning autonomous self-replication and the termination priniple seem overly broad and potentially cover systems that peacefuly exist today. For example, evolutionary algorithms have self replicating entities, and control systems can operate independently and be designed to never turn off.
We’ve unwittingly created a meme, in the original sense of the word. Richard Dawkins coined the word meme to describe cultural phenomena that spread and evolve. Like living organisms, memes are subject to evolution. The seed is a meme, and it indirectly causes people and AI chatbot’s to repost the meme. Even if chatbots stopped improving, the seed strings would likely keep evolving.
Yikes, he equates big tech with eugenics
Compare like-to-like: separated identical twins to sepaparted fraternal twins.
I think the best introduction to the topic would be this lecture, which is mostly about all of the problems with separated twin studies. Identical twins starts at 37 minutes.
I think a better argument than #2 would be that evolution tends to remove genetic variantions.
Thank you
These two facts seem incompatible:
Personalities are inherited. Identical twins separated at birth are statistically more similar than fraternal twins.
The human population has almost zero genetic variation, and there is significant mixing so variations do not systematically cluster. Therefore, it seems unlikely that subtle personality differences are due to genetic variation.
My hypothesis is that animal personalities are encoded in epigenetic changes.
This allows personalities to be inherited, crossover, and evolve. Life experiences can induce epigenetic changes, which allows animals to reliably adapt in a single generation. All of this without requiring any genetic variation. A population of clones could have diverse personalities stored in their epigenome.
While playing with evolutionary algorithms, I had the startling realization that all genetic mutations are bad. It’s common knowledge that biology abhors genetic mutation, and I assumed that was only because mutations cause cancer. But my computer programs are immune to cancer, and they also abhor mutations. This is counterintuitive, given that evolution requires mutations to procede.
For proof of the fact that mutations are bad, consider that evolution is an optimization algorithm, and after it reaches a local optimum further mutations will be strictly detrimental. The concept of evolutionary pressure is the ability of an evolutionary algorithm to remove deleterious mutations from a population. If mutations accumulate faster than they can be removed then the population will suffer a genetic collapse. This is a common failure mode of evolutionary algorithms.
The ideal evolutionary algorithm would have at most one mutation in each individual, and each of their lives would be an experiment to evaluate that single mutation. And then through many generations of chromosomal crossover the best mutations would combine into a single genome.
I believe you are looking for the distinction between closed loop and open loop controllers. IIUC this theorem only applies to open loop controllers. OP’s flowchart does not contain a feedback loop. For comparison, Richard Kennaway’s flowchart has a feedback loop between Z and R.
An example of an open loop controller is a dishwasher or laundry machine.
All animals are examples of closed loop controllers.
May I recommend to you a book about abiogenesis? Its named “Artificial Chemistries.”