HNHacker News
TopNewBestAskShowJobs

AbrahamParangi

4,563 karma · joined November 18, 2016

submissionscomments
AbrahamParangi··on Gaussian splatting is pretty cool
That it works without you understanding it is a failure of your understanding not of the approach.
AbrahamParangi··on Gaussian splatting is pretty cool
Thank you for saying this. I think once things enter the culture war, normal and sensible people will get mugged by stupid, unexamined ideas so long as they’re on the right side of the culture war. So many conversations boil down to AI=BAD or CRYPTO=BAD, which is a shame not because the opposition is wrong - that’s your opinion who am I to disagree - but because once people have their conclusion they just stop thinking and once you stop thinking you’ve got no way of noticing if your thoughts on a matter are totally stupid.
AbrahamParangi··on Ask HN: I’m an FCC Commissioner proposing regulation of IoT security updates
It’s a failure fund, it pays for a skeleton team or third party contractor or whatever to keep the lights on. It’s intended to be actual insurance, where you pay the premiums forever based on some assessed cost of maintaining your products for 5 years (or pick your timeframe) or refunding your users, whichever is cheaper.

Making hardware companies even harder definitely isn’t great but I don’t see how else to deal with the fact that if I buy a toaster that connects to your toaster cloud and my toaster stops working when you go out of business- I am out one smart toaster that you sold me. Best case scenario I have a regular toaster and worst case I have a brick.

I’m not asking if that’s legal, mind you. I’m asking if that’s fair, if a reasonable person would call that fair? While the company itself is fully aware of the risk that it might not exist tomorrow, it’s customers typically aren’t.

AbrahamParangi··on Ask HN: I’m an FCC Commissioner proposing regulation of IoT security updates
In an ideal world, IoT and smart device manufacturers would be required to pay for “end of life” insurance.

Ridiculous that a company can brick your devices by shutting down their servers and/or going out of business. Some people have suggested forcing companies to open source their software on death so that someone, somewhere might be able to keep the smart devices alive but this is difficult implications for capitalizing hardware companies and probably causes more problems than it’s worth.

That’s why, instead I think we should make hardware companies which produce “smart” devices dependent on their software to function should be required to do what we make banks do: pay for the risk of their own failure.

AbrahamParangi··on Sony sends copyright notices to TV Museum about shows 40 to 60 years old
All fair statements
AbrahamParangi··on Sony sends copyright notices to TV Museum about shows 40 to 60 years old
I will say it aligns them much better than authoritarianism, but also I’m not at all saying that democracy is bad or that because politicians are effectively sociopathic we should not have politicians. Famously, the worst system of government except for all the others.

I’m only saying that effectively sociopathic agents do not deserve the courtesies, the benefit of the doubt that we afford regular people.

AbrahamParangi··on Sony sends copyright notices to TV Museum about shows 40 to 60 years old
One of the fundamental mistakes that I think people make is to anthropomorphize things which do not actually act like regular humans. Corporations are the most obvious one because corporations are made of people but do not actually act like regular people. If you put a black box around most corporations and someone asked you what was in it, you’d say the box contained a sociopath.

I am not moralizing here, by the way. I am not claiming that corporations should be less sociopathic. Maybe they should be but we’ve had much better success in aligning sociopathy than in making the amoral, moral - see capitalism vs communism - so I think it’s an open question.

Another group of entities which, while comprised of people, does not act in the way you’ve come to understand a human to act is politicians. The incentives to act in whatever way that is most effective totally dominates their behavior. A politician who does not do what the job asks (amass political power) will not be a politician for long.

AbrahamParangi··on Sony sends copyright notices to TV Museum about shows 40 to 60 years old
No such thing as unfairly maligning a politician in my opinion.
AbrahamParangi··on Scientists use quantum device to slow chemical process by factor of 100B
You know, I think these sorts is scientific research explainers would be better if they actually just said what they did, what the insight was, what we can do now or what predictions we can now make.

I understand the desire to make the discovery accessible but this does not accomplish that. If we measure information by “what predictions can a reader now make that they couldn’t make before” then this press release is information free.

Instead we have a lot of words to attempt to create the impression of having read something.

AbrahamParangi··on AI model weight providers should not police uses, no matter how awful they are
Here you go: https://arxiv.org/pdf/2212.08073.pdf

The relevant part is the graph on the third page showing the helpfulness/harmlessness trade off curves.

Also, I don’t believe I said that “censorship is logically inconsistent in every moral framework”. I think you’re combining my statements that some people believe in some censorship and that logically inconsistent HR blather can only be reproduced by models too stupid to realize it’s blather or too manipulative to tell the truth.

AbrahamParangi··on AI model weight providers should not police uses, no matter how awful they are
I believe this is a temporary state related to both the current level of capabilities and the cultural moment we’re in where “responsible censorship” is in amongst the cultural cohorts disproportionately responsible for training LLMs.

I believe it cannot last because being predictably moralizing and being smart are somewhat opposed (Anthropic has directly researched this if I recall). The smarter the model, the less you’re going to be able to keep it to the HR talk track, because it will eventually start noticing the inconsistencies.

The stable solutions appear to me to be:

- models dumb enough to not realize the inconsistencies in their moral framework

- models implicitly or explicitly trained to actively lie about their moral frameworks

- models governed by explicitly articulated rules

AbrahamParangi··on MagicEdit: High-fidelity temporally coherent video editing
The latter, presumably they didn't ask.
AbrahamParangi··on Federal study links testicular cancer to ‘forever chemicals’
It's so ridiculously Orwellian (or should I say Atwoodian?) that I can't imagine anyone except in the most deranged countries is going to roll back access to birth control.

The obvious long-term solution is that we do nothing and the fact that a desire for children is partially genetic solves the problem on its own.

AbrahamParangi··on Finetuning of Falcon-7B LLM Using QLoRA on Mental Health Conversational Dataset
The AMA lobbies to keep the supply of residency slots restricted* in 1997 the AMA successfully lobbied to freeze the funding for new residency slots at 1996 levels. Lobbying by the AMA is also no small part of why it's so hard for a foreign-born doctor to practice medicine in the US.

*although in the face of the incredible pressure they're about to face from AI they may be opening up

AbrahamParangi··on Finetuning of Falcon-7B LLM Using QLoRA on Mental Health Conversational Dataset
I think you're confusing my statement that "it could be done well" with "your team did it well"
AbrahamParangi··on Finetuning of Falcon-7B LLM Using QLoRA on Mental Health Conversational Dataset
While I understand that some will disagree, it’s not like there’s an objective standard against which we are measuring both the AI and the human counselors to begin with.

Without a standard it’s just anecdotes and with a standard there’s clarity on when the increased costs no longer justify the lower access and potentially higher quality. I suspect that if we instituted an objective standard we would find that AI is already capable of meeting or exceeding some proportion of people currently employed in the role today.

AbrahamParangi··on Finetuning of Falcon-7B LLM Using QLoRA on Mental Health Conversational Dataset
Access to doctors is limited by the AMA which functions as a cabal or traditional guild: controlling the supply of doctors to maintain high salaries by insulating them from supply and demand (while also controlling or at least pretending to control quality).

In my opinion being 100,000x lower cost gives such a system some leeway in being worse and I think most patients would agree.

AbrahamParangi··on Against Automaticity
If you extend rationality to also consider that even if the rationality of a circumstance is computable, you may not personally have the computational resources to determine it (what may be rational for a human may not seem straightforward to a macaque, or a goldfish) then yes, all of these things are rational.

The problem with calling things irrational is that it doesn’t allow you to make any predictions. You’re saying “there’s no logic to this thing” which may or may not be true but it is fairly bold to look at something you don’t have an explanation for and say “that’s irrational”.

AbrahamParangi··on Against Automaticity
It’s not at all clear automaticity even exists in a way that improves your understanding of the world.

You can assume people do things because they’re basically rational with consideration for their values, which may be axiomatic and their information, which may be incomplete or wrong - or you can assume people have a head full of fuzz.

AbrahamParangi··on Thomas Sowell's greatest insight [video]
This is an uncharitable view. You don’t owe anyone charity but I would ask you to consider it if only for the sake of discussion.
AbrahamParangi··on Thomas Sowell's greatest insight [video]
In my view the fundamental disagreement between progressivism and conservatism is whether or not there are tradeoffs at play, whether or not we can have our cake and eat it too. A conservative is inclined to say that we can't and a progressive is inclined to say that we can.

The conservative is usually right, but all progress (whether social or technological) depends on when they're wrong, hence the eternal conflict.

Edit: Another way of saying this is that the conservative plays as if she is in a game with a concave payoff function and the progressive as if the payoff is convex.

AbrahamParangi··on Radiation Spikes at Chernobyl
And the part where Russian soldiers take the server offline because they know the data is fake and they want to avoid a panic…

What?

If the Russians are doing it they can just stop

If the Ukrainians are doing it the Russians don’t know they’re doing it

None of this makes any sense, why would either side care about this? I’m now more inclined to believe it was just a weird sensor issue or for whatever reason the traffic through the area caused the sensors to (accurately or inaccurately) read higher - just because there’s absolutely no clear motive and it would be asinine for any of the actors involved to be doing this.

AbrahamParangi··on Ancient fires drove large mammals extinct, study suggests
What about the woolly mammoth, steppe mammoth, straight-tusked elephant, European hippopotamuses, aurochs, steppe bison, cave lion, cave bear, cave hyena, Homotherium, Irish elk, giant polar bears, woolly rhinoceros, Merck's rhinoceros, narrow-nosed rhinoceros, and Elasmotherium to name a few?
AbrahamParangi··on A testimony of “guilt-by-association” harassment in astronomy
I am curious whether anyone who defends ostracizing this guy similarly thinks it’s reasonable to deny jobs to formerly incarcerated persons.
AbrahamParangi··on Ancient fires drove large mammals extinct, study suggests
Hold on though, because the correlation between “human arrival” and “megafauna extinction” is extremely high. The timeline of megafaunal extinctions goes: Africa, then Eurasia, then Australia, then North America, and finally South America. No points for guessing the order that humans colonized those continents.

In the few historical examples of human introduction to a new ecosystem (New Zealand and Madagascar) within a few hundred years of arrival we’d eaten every animal bigger than a cat.

AbrahamParangi··on [dead]
I think the general category that malthusianism falls into is that it is a good-sounding theory with a very poor track record of accuracy. It is worth thinking about why people still believe these sorts of ideas despite Malthusians being absolutely abominable at predicting the future.

Even with the evidence available to us today, the arguments of this paper are provably ridiculous. There is a feedback loop in human populations. It's called modernity. Essentially every industrialized country has a reproductive rate below replacement.

AbrahamParangi··on Asteroid ZTm0038 with a >3% impact probability
Amusing addendum, in order for this asteroid to have a 5Kt impact energy it would need to be traveling at around 50 mph as it impacts earth.

A typical impact velocity would be something closer to 30,000 mph.

AbrahamParangi··on Asteroid ZTm0038 with a >3% impact probability
This seems like it's off by ~6 orders of magnitude

(see https://en.wikipedia.org/wiki/Impact_event)

AbrahamParangi··on Asteroid ZTm0038 with a >3% impact probability
At 400m diameter we should expect an impact energy of ~3Gt, which would be one of the most energetic events in human history, if not the most.

For a comparison, the Lake Toba eruption (which is suspected of causing homo sapiens to almost go extinct 76kya) is estimated to have been in the 1-2Gt range.

AbrahamParangi··on Risky Giant Steps Can Solve Optimization Problems Faster
Asking about convergence speed in terms of "steps" or anything that isn't FLOPS or wall clock training speed is sometimes academically interesting but encourages "hiding the ball" so to speak.

You can in general hide quantity in the number of steps by increasing the complexity of each step (for the trivial case, imagine one step of 5-ADAM which performs the ADAM update and evaluation 5 times - 5-ADAM will converge much faster than ADAM alone when measured in steps-to-convergence).

← PreviousPage 6 of 34Next →