nomadpenguin

It's a genuinely fascist tendency. Only the elite few are "real people" who think. Everyone else is irrelevant, democracy is a failed experiment, wealth = worth, etc etc etc

Yeah, this smells very fishy to me. It's almost trivially easy to gather a small validation dataset in humans for the paper. At my institution, it's about $800 an hour to scan someone. You can probably get enough data to validate the model with a half hour scan. Surely the group has enough grant funding to pop a few healthy controls in the scanner.

I haven't looked in super close detail to the paper, but their methods section says that they fit a video model (V-JEPA2) to the fMRI dataset in a voxelwise ridge regression, meaning that the baked in assumption is that the visual response affects each voxel independently. Voxelwise models are very nice for making statistical inferences, but are less good for prediction and modelling tasks, because our brains certainly do not work as collections of independent regions.

BOLD is intensely messy data, and their design is far too simple IMO to reflect anything of reality.

For me, the biggest AI writing tell (other than the blatantly obvious ones) is an unnatural consistency in style, whatever style that may be. It's most apparent in longer pieces, and I'm not sure I can really pin down exactly what it is. But human writers seem to lack the ability to keep a 100% consistent voice and lapse into different registers at different times. LLMs don't have this natural rhythm, which makes for an exhausting reading experience.

The Cincinnati Zoo covered all their parking lots with solar panels last year. Your car stays cool in the summer, and there's motion activated lighting under the panels after dark. It's awesome.

"Genetically encoded" is the appropriate term here -- it was used in the original journal article. It's a common industry term in neuroscience research. For example, GEVIs and GECIs are "genetically encoded voltage indicator" and "genetically encoded calcium indicator" respectively. "Genetically encoded adenosine sensor" here is a term of art.

I have over 14 years of education in developed countries, and out of those, maybe 1 year combined meaningfully helped me in my jobs/career in terms of skills.

I think you're underestimating the effect of 14 years of daily training in literacy and numeracy.

I believe Macrofactor has had these features for quite a while now.

I think Buddhism still (arguably rightly) doesn't sit entirely well with non-religious Westerners. I have studied with a Zen Sangha and transmitted teachers on and off for a bit and have found their explanations helpful. However, it's absolutely undeniably that the Buddhist cannon is full of batshit insane stuff, just like any other religion. You can write them off as skillful means, but in some ways I think it's more honest to say that you practice meditation with Buddhist characteristics than to say that you're a real Buddhist if you don't have the time of day for spirits and dieties.

Again, this isn't saying that Buddhist modernism is bad. I'd argue that having clear eyes about what parts of Buddhist practice you're willing to take and leave is good.

There are specialized architectures (the Tolman-Eichenbaum Machine)* that are able to complete this kind of task. Interestingly, once trained, their activations look strikingly similar to place and grid cells in real brains. The team were also able to show (in a separate paper) that the TEM is mathematically equivalent to a transformer.

* https://www.sciencedirect.com/science/article/pii/S009286742...

As Freud said, there is no negation in the unconscious.

He also believes that he can remember full memories from when he was 3 months old. There is no evidence that this is possible.

How are you handling scheduling with FSRS? The challenge that I quickly saw was that it was difficult to figure out when you should advance a segment of information. If you get 80% of the info right, should it be advanced? What happens to the 20% you missed? How do you prevent yourself from missing the same 20% every time it comes around?

Poor generalization (overtraining on prompts) and loss of context over time are the biggest issues I've found with them. Slow card creation workflows and needing to rate your own reviews are merely UX issues -- losing context and losing generalization make SRS actively harmful when used for some topics.

There's 2 solutions I've thought of but haven't tried implementing:

1. A free-recall based approach. Free recall allows you to operate at a higher level of organization and connect concepts at lower levels. However, how you would schedule SRS with free recall is not clear.

2. Have an LLM generate questions on-the-fly so that you don't overtrain on prompts. You might also instruct the LLM to create questions that connect multiple concepts together. The problem with this approach is that LLMs are still not so good at creating good test questions.

Yes, that is true. I phrased that badly -- it's more that we didn't take the evolutionary branch where we retain the fetal hemoglobin because it is maladaptive in adults.

High affinity RBCs would actually be a disadvantage for athletics. You actually don't need very high affinity to pick up oxygen from the lungs -- your lungs are comparatively extremely high in oxygen. What matters more is being able to drop the oxygen off in peripheral tissues. Higher affinity means that it's harder to actually deliver the oxygen, which is why we evolutionarily developed the switch away from fetal hemoglobin.

This is correct, it's called empiric treatment. If a patient comes in with altered mental status and neck rigidity, you don't have time to take a lumbar puncture and culture bacteria. I don't know anything about phage treatment, but from what the other commenter said, it seems like then you'd have to do some sort of PCR test as well. You simply don't have time for any of that -- your only choice is to blast them with vancomycin + ceftriaxone.

You're using two en dashes to approximate it -- few people have the en dash character on hand.

No amount of willpower or happy thoughts will cause your hormones or neurotransmitter levels into the homeostasis you want.

Behavioral-only therapies are effective, even if they're not as effective as we'd like. We have the ability to modulate our own chemistry.

What do you mean by loopable mindmap? What software do you use for this?

Which is itself a major risk factor for dementia (and strokes)

That would suck. I want to see something I haven't seen before.

The catch with deep brain stimulation currently is that it's only SOTA for implanted electrodes, meaning it's incredibly expensive, and while DBS implantation is very safe for a brain surgery, it's still a brain surgery. tTIS is very investigational and neuromodulatory focused ultrasound is still quite a bit away from clinical applications. (Ablative focused ultrasound is FDA approved, but that's only for very specific indications.) There's also the Brainsway H series TMS coils that claim to stimulate deep structures, but the activation of large amounts of cortical tissue makes the claim a little hard to verify.

Why do you think Tantacrul is seedy? I've watched a couple of his videos and thought they were entertaining; does he have a history of doing shady things?

I think there's a huge division in toxicity between wet bench and computational/dry lab work. If your math PhD friends go on a wild goose chase or slack off for a few weeks, the only thing that is lost is time. In a wet lab, it could cost you hundreds of thousands of dollars. The stakes, stress, and constant attention required from experiments feeds into the toxicity in wet labs.

Adding to that, a large portion of (important) wet bench work is mind-numbing manual labor. This work really should be done by a tech, but techs are just as if not more expensive than grad students, and techs can leave the job if they're not happy. Which means that grad students are at the bottom of the totem pole of intellectual respect, and PIs who had to do the same expect a lot more "due paying".

Meanwhile, students in computational labs are working from home.

Perhaps all the sleep allows them to be energetic.

I've been experimenting with "spaced free recall". So first, I'll read a section of a textbook. Then, I write down everything I can remember about it in a blank text file, organizing things in a way that makes sense to me. Next, look back at the section and compare to my recalled notes, filling in missing information and committing extra attention to missed spots. Repeat the process with increasing intervals between reviews.

From what I understand of the literature, free recall produces better learning compared to cued recall like flash cards. Part of the reason is that it forces you to organize information and associate it with existing knowledge.

Anecdotally, it's much easier to learn conceptual knowledge, and I don't really feel like my recall of specific facts has suffered compared to traditional SRS.

I recently worked on a project involving segmenting clinical CT scans. Almost every python package for working with CTs was written by the same guy.

Good to see that the Duke meme god is still meming

An important piece of context is that many of the New Left thinkers you mentioned are responding directly to the horrors of WWII and the Holocaust. While yes, tribalism, brutality, and oppression are all present in non-Western cultures, I don't think it's an absurd claim that it is only Western rationality that can produce mass suffering at the scale of the Holocaust.

"Bentham's panopticon reified" is more characteristic of Foucault's concept of disciplinary societies, which defined the 20th century. Deleuze's project here is to extend it into the 21st century, where the disciplinary society changes into "societies of control". While both disciplinary and control societies rely on surveillance, disciplinary societies create compliance through the threat (if not the execution) of punishment, while control societies create compliance through nudges and dark patterns. Disciplinary societies have centralized locations of power (the factory, the school, and the barracks), while in control societies, power is decentralized -- both everywhere and nowhere at once (algorithms, workplace encouraged education and health initiatives, etc).

Furthermore, while disciplinary power creates classes of individuals to control and marginalize (based on race, class, etc.), under neoliberal control societies, all individuals are prima facie accepted. This is not because the control society is less coercive, but because control is exerted at the sub-individual level -- think how advertising algorithms assign each person a constellation of tags which are used to structure their online experience.