abeppu

39 ms·

511.
by abeppu 2y ago
> EEG - 10 channels dry, soft-polymer active electrodes > EMG - 4 channels > EXG - 2 channels > EOG - 2 channels I thought "EXG" was generally used to indicate "EEG, EMG, ECG or EOG" ... but in the context wh
512.
by abeppu 2y ago
... is that a good example? I think people who use the dating apps today mostly hate them, and find that they have misaligned incentives and/or encourage poor behavior. There's been generation of other services that shift in popul
513.
by abeppu 2y ago
> Not sure if meeting rooms are critical part of RTO efficiencies in anyone's mind. No, it's not the point people bring up -- they like to talk about being able to just walk over to someone to ask them a question, or quickly fo
514.
by abeppu 2y ago
I don't doubt that this happened, but what percentage of RTO efforts were about intentional attrition vs something else? The intentional attrition strategy seems like it only makes sense if you have a lot of dead weight on your staff b
515.
by abeppu 2y ago
... I do think "it's unstable" is kind of a non-answer though. Lots of control problems are unstable, right? For example here's a successful project from 16 years go where Andrew Ng, Pieter Abbeel etc used apprenticeship
516.
by abeppu 2y ago
But it's interesting to see that this is not true for all crime categories. E.g. car break-in hotspots are by Fisherman's Wharf and the side of Alamo Square with the Painted Ladies -- i.e. prime tourist locations. However, I sus
517.
by abeppu 2y ago
It was kind of disappointing to see the transition from fancy, glamorous shots of it standing still and looking around/at its hands to plodding slowly and cautiously (I was kind of reminded of walking in ski boots?) across a flat, open
518.
by abeppu 2y ago
I don't think it's just confirmation bias where we ignore some bad results (which presumes we know up front that they're bad) -- I think because these models are specifically RLHFed to learn what we think looks good, you ca
519.
by abeppu 2y ago
... and I guess another question is, how stable is this trait? E.g. if we got really used to telling 12 year olds that the marshmallow test finding indicates that the ability to put of immediate rewards for larger later rewards is really i
520.
by abeppu 2y ago
I think a weird irony is that the model's inability to know when its response is good is both the reason why often the output is not useful, and why when it's very useful, they can't capture the value efficiently. Like, I was
521.
by abeppu 2y ago
Aside from tgv's correct point that this is implicitly a recipe for something that isn't useful as a visualization, I think even if we were able to distinguish 4B colors and make sense of each pixel -> color assignment ... the
522.
by abeppu 2y ago
And the examples given here can make the walking experience better, for a similar amount of total expenditure, without meaningfully changing the situation for drivers. Some things the author never suggests in this post: - removing lanes of
523.
by abeppu 2y ago
... but what are "these posts"? Because this post compares good and bad examples within the Minneapolis-St Paul metro area. This isn't a comparison of some cherry-picked European city with the rural US. It's a comparis
524.
by abeppu 2y ago
I regularly walk by one of their parking lots (14th St & Shotwell), which used to be typically relatively full, and now it's generally pretty empty. They're all out and about ... but on the other hand, a significant proportion
525.
by abeppu 2y ago
1) But if it's just an avatar chatting on a screen, then this only works for people in the specific window where they're motivated to stay engaged with the chat. I'm not in elder care, but when one of my family members had de
526.
by abeppu 2y ago
It's interesting to me that this page doesn't mention "safety" or content moderation, etc. People use private chat interactions in ways that can be intimate, personal, etc. I'm guessing that for Meta, policies about
527.
by abeppu 2y ago
... but the article describes that the antibiotics being discussed combine structures from classes of antibiotics which are already used independently. > Macrolones are synthetic antibiotics that combine the structures of two widely used
528.
by abeppu 2y ago
No, bad/wrong/nonsense is not the only risk here. You're missing the main point that the authors are making: the shape of the distribution gets changed by this process. A model trained on human data will produce fewer high-
529.
by abeppu 2y ago
Yes -- said another way, if you're an ML researcher and you have human-provided (scraped) data, and an ability to generate synthetic data, then until recently, you had a controllable parameter: how much of your training data for your n
530.
by abeppu 2y ago
... but a bunch of companies were, and a lot of the claims about what web3 was going to be were really vague
531.
by abeppu 2y ago
Is this true in Germany? As an American, I was always of the impression that in continental Europe, tips were as a rule a much smaller part of compensation.
532.
by abeppu 2y ago
> Over the past eight years, scientists from Sheffield and COMSATS University Pakistan have been studying how the sugar can help to heal wounds by promoting the formation of new blood vessels. During the research, the team also noticed t
533.
by abeppu 2y ago
I'm not sure that we disagree, but perhaps we're understanding "launching" in the metaphor somewhat differently. I think the key point about the interstellar probe example is it's assumed that once it's launche
534.
by abeppu 2y ago
> It's not enough for a real tuition or to support them to study instead of work. ... should it be? Or should you have to save some of your basic income for a period to go to school? If every year you got enough to live off and to b
535.
by abeppu 2y ago
I mean, I love a tax cut too, but I don't think taxes per se are the reason why people aren't having kids. This data is somewhat stale, but here's a graph of taxes as a % of GDP vs birthrate. Color and size are both tied to i
536.
by abeppu 2y ago
I saw some blurb (I can't find it now) about the optimal time to launch an interstellar probe, and the punchline was that embarking too soon means you will be overtaken by a later vehicle which launched later but with better tech. When
537.
by abeppu 2y ago
It's a good point that whether something is testable changes over time. But, in separating scientific from non-scientific claims, can we not distinguish those that are in some way motivated by some limited available evidence, and which
538.
by abeppu 2y ago
Strictly, I think your first sentence is correct: pseudoscience sometimes does make falsifiable claims. Homeopathy advocates believe that their preparations actually help heal. But for your second sentence, do you have in mind a hypothesis
539.
by abeppu 2y ago
> internal locus of control Can you expand on that? It seems potentially but not necessarily consistent with "democratic institutions" but the "control" must be limited to preserve the various freedoms required for op
540.
by abeppu 2y ago
I mean, the original article doesn't say anything about video models (where, frankly, spotting fakes is currently much easier), so I think you're shifting what "they" are. But still: - input doesn't distinguish what

More ›