Prometheus

Ah, you want to statistically compare time series. That sounds tricky, though a quick google shows it might be possible, but i can't help you there. I would ask whether you really want to compare trends over time though, which is what you're doing with time series, or are you only care about comparing the means of the pre and post groups then more standard stats test will suffice.

Some quick thoughts. Paired? That implies you are able to match individual criminal's pre and post scores. If so, great. You need to start looking at the distribution of your data. It's count data that has been normalised - i'll bet you it's not normally (or Student T) distributed. You may also suffer from an inflated number of zero counts (i.e. some individuals just don't use some words that others do). What are the actual outcomes here? Presumably you have many keywords, therefore its not a single outcome but a collection. Or do you pool all these results into a single outcome somehow? Best bet is to consult a statistician who will tsk at you for talking to them after you collected the data. Next best is to delve into the literature and see what other researchers have done - particularly if there's a statistician on the paper, they will likely have encountered many of the problems you will.

I've not heard that saying before, but i hear about unsafe roads all the time. Everyone seems to know a dangerous stretch of road, a death trap - such roads are often modified to make it safer, if even that's just speed cameras. I don't think it unreasonable to modify roads slightly if it helps automated vehicles out, though constrained by financial realities. If, for instance, the particulars of the cattle grid and the hill that combined to cause issues in the BBC article occurs just a few times in the world, then changing the infrastructure seems reasonable. If, however, it's every other cattle grid then changing the vehicle seems more reasonable.

It is irrelevant whether Sharia law is 'compatible' with Western law or not: you cannot have a single country that runs one system of laws for some of its citizens and another system of law for others. That way lies anarchy. This is known as the no true scotsman fallacy - rather than actually confront why Islam can be interpreted in nefarious ways, the problems are just waved away as not being 'true' Islam. Of course, they will say the same thing about your interpretation of Islam.

Don't know. You've read through their accompanying paper? Under their FAQs is a contact form - i'd ask them directly.

From their website: CDR and dementia diagnosis can be found with other demographic data in “ADRC Clinical Data”. Or you take the opportunity to try some unsupervised learning.

Isn't that more a definition of machine learning, which is itself a subset of AI?

Even if they prove correlated, with millions (billions?) of astronomical bodies now being tracked our sun is bound to be in sync with some of them for some of the time by some metric.

Does spatially-resolved have a specific technical meaning in this context? Spatial information is acquired by taking a spectrum from a defined location then shifting the stage slightly and taking a spectrum from an adjacent location and so on, then computationally stitching them all together.

Hyperspectral imaging using Raman microspectroscopy is a thing, the putative benefit being that images of various biochemicals can be acquired without the need for individual staining, as is currently done via immunohistochemistry. It's quite a specialised setup though. I gather the OP speculates such techniques can be modified to allow brain imaging? If only. I've seen some transcutaneous experiments, mostly for blood glucose monitoring which isn't concerned with spatial information. Going through skin is one thing, i've never come across any set-ups even trying to penetrate bone. I don't know physics, but one problem that straight away suggests itself is that if you are using longer wavelengths to achieve penetrance, then you are limiting the usefulness of any spatial information (as there is a dependence on wavelength and spatial resolution).

This collapses down to a value alignment problem. As a super-intelligence it should be able to predict people quitting their jobs etc. Whether it knows this is not what we really want depends on its goals. Is it simply maximising dopamine - then it could invent a way to directly stimulate dopamine receptors. Is it trying to cater to every physical whim - then we could end up with enforced hedonism. Is it trying to to optimise for some vague concept such as 'wellness' - this might sound ideal as wellness could include just enough resistance for us to overcome to make human life fulfilling, but can we define such vague goals? There are attempts to have AI agents that extract their goals from the environment instead of having them explicitly stated, which may provide one avenue to this end.

You mean SN8? I'm not an engineer but my understanding is that it hit the ground too hard as they couldn't get enough propellant to the engines. It crumpled where it was most structurally weak. I think they regarded it a resounding success. SN9 might give it another go this week or next.

Yes. Again, it depends on the culture. I can imagine something even as simple as ubiquitous automated driving would be well received and readily adopted in somewhere like Singapore, but rejected in many places in the US. Scale that up. I'm not sure that is true: most religions are predicated on the observation that on some level humans are not in control. Whether that manifests as a god/s being in control, or natural forces (to which humans are bound) is irrelevant - the idea exists in many systems of thought. Perhaps you mean human agency? I don't see that humans would necessarily give up this agency in light of a super-computer. Computers already play chess much better than all humans, but AFAIK that hasn't affected the numbers playing chess in the slightest. Nick Bostrom gives an account of this in his book Superintelligence in which he outlines several paths superintelligence could emerge and speculates that the most destabilising ones are one that emerges alone (i.e. the Chinese of Americans get the first, and so only superintelligence, as it can destroy all other attempts) and/or one that emerges so quickly that societies cannot react, either internally (emotional) or externally (putting in place laws) - we already see how slowly governments are responding to social media. The other scenario he warned of was an arms race to super intelligence in which AI safety (value uploading, goal misalignment, orthogonality thesis etc...) are ignored just to beat the competitors - which, i believe, is why open AI was founded.

I find anaconda useful for managing environments. I also had problems installing tensorflow, then a bunch of computer scientists said google didn't support it that much and that Pytorch was the way to go anyway.

What's the purpose of the project? Are you focused more on learning how programming a chatbot works, or do you just want a working product (or something else)? The full model has 175 billion parameters. Crazy stuff. I heard rumour that GPT 4 will have 20 trillion parameters. How much do you think just making bigger models and feeding them more data will improve outcomes?

The Markov model linked isn't trying to extract intent, i think it's predicting the next word in a sequence given an input word, based on a state space that contains emission probabilities for word pairs. It then uses that output as the input to predict the next word until a termination is paired with an input word. Intent is how humans parse such information, there's no reason an algorithm needs to parse words in the same way (and still give intelligible results). If you want state of the art natural language processing check out GPT-3. It's a neural network based on an autoregressive model (so that it can take into account a number of previous words rather than just one), but i believe it still just predicts the next word sequentially.

It may be possible that science, through technology, can provide a post-scarcity society, where all our physical needs are sated. Arguably some people already occupy that bubble. Whether that counts as a utopia or would be sufficient to create the conditions of a utopia is another question. I'm reminded of the rat utopia experiments in which rats were provided every physical need and didn't do so well. I suspect humans would fare little better. Or as William Blake put it: Man was made for joy and woe Then when this we rightly know Through the world we safely go. Joy and woe are woven fine A clothing for the soul to bind.

A major consideration would be whether your data comes from a truly random sample - given the nature of what you describe, probably not. This will limit your options, but there are plenty of observational study designs you could utilise depending upon the exact nature of the study and what question you want to answer.

That's exactly what insulin does - without insulin cells cannot absorb glucose through facilitated diffusion thus blood glucose levels accumulate. Giving insulin treats this. Unless by treat you mean only therapies which would restore function to the pancreatic beta cells which fail to produce insulin in type 1 diabetes. This is not currently possible. Perhaps there is a language barrier here.

Two cases is not enough to draw the conclusion that 'medicine treats symptoms rather than disease'. It's true that there are cases for which medicine can only treat symptoms - these usually reduce to not understanding a disease sufficiently. But there are plenty of cases where the cause of the disease is directly addressed - antibiotics, for instance. Even your own example of diabetes is not representative - all type 1 and many type 2 diabetics take insulin , which is not treating a symptom. I didn't understand the points about hypertension, melanin or skull shape.

Appreciated, though there are still a few teachers i would slap if ever i met them again. I think the thing that pissed me off the most though was that my GCSE science teacher had never even heard of the big bang, and laughed at me when i asked what exotic particles are.

I think that's an unfair representation of what MigL said, at least i didn't read it that way. Lack of motivation =/= laziness. You could be an extremely hardworking and unhappy warehouse worker, but never even think you could improve your lot through education. It's just not a thought for many socio-economically disadvantaged people. If your dreams provide the ceiling of what you can achieve in life, then poor people in rich countries are conditioned to dream no higher than the dog's bed. It's as much a barrier to education as is anything else, but perhaps the most important because it's the only one you can directly tear down with your own mind. That's not mutually exclusive with making changes to education system, but while we're waiting for that to happen, as Billy Bragg says: the system might fail you, but don't fail yourself.

I think this joke perfectly captures the essence of the topic. In this joke the monks were just following teachings blindly. Apparently they never questioned or explored their celebicacy. But i understand (some?) monks are encouraged to explore the experience of celibacy. It's not just an arbitrary rule, but a tool used to explore a headspace few humans choose to navigate. There's an inquisitiveness to it. In this case the transcription error wouldn't matter to their practice because they are focused on the experience. The mindset of the former monks might not be conducive to science, but the latter monks would have an easier time of it. Now it might be that certain religious institutions encourage one way of thinking over the other, but religions are not homogenous and each should be taken on their own merits.

I remember a conversation with a Pakistani friend at school where he said he was going to try to become a doctor and that i should do the same. I just laughed at him - i've no idea where i picked it up, no one had ever explicitly said i couldn't, but even the idea of being a doctor was already beyond me. I was, however, explicitly told by my teachers that i couldn't be a pilot or a scientist.

I don't think it's an unreasonable question, after all astronauts spend some of their training under water to better approximate zero-g conditions. Unfortunately are probably not a good model organism in this context: