varelse

70 ms·

391.
by varelse 5y ago
Or they have a story far more complicated than what have you have reduced them to being. Just attempting to manage the pandemic in the US has demonstrated how complicated and contradictory people turn out to be. But I do agree that pamphlet
392.
by varelse 5y ago
You're using outliers to invalidate far more accurate predictions of the ensemble. You're right about people like yourself. IMO they are not the norm. Most "Weekly World News" readers are, I suspect, far more gullible th
393.
by varelse 5y ago
I suspect if they provided telemetry into the Facebook Zeitgeist, it would show its userbase to be terrifyingly stupid and superstitious folk on par with the late Carl Sagan's predictions in _The Demon Haunted World_. For IMO the Googl
394.
by varelse 5y ago
We agree that the incentives for startups are completely broken right now. But until that environment changes, the bar is pretty high for disrupting current methods.
395.
by varelse 5y ago
Which is why you need a killer demo. New ideas are great, pure research rocks. But new ways to solve real world problems derived from that research is the killer product IMO. See an unnamed AI ASIC company proudly announcing they have achie
396.
by varelse 5y ago
You might want to watch Nvidia's presentation on the recent MLPerf competition. There's a tremendous amount of engineering going into making this stuff more efficient on current hardware. https://developer.nvidia.com&#x
397.
by varelse 5y ago
IMO these sorts of methods become interesting the day they are demonstrated working in a conventional AI framework on a commercially important graph like Resnet or BERT. Everything else follows from a killer demo like that. The benefit is o
398.
by varelse 5y ago
I've done OK avoiding the brainfarts of thought leaders and other fools so far. But I grew up surprisingly deep inside drug culture with 2 drug-dealing siblings and their drug-dealing shenanigans. Yet I believe there's nothing wro
399.
by varelse 5y ago
My bad: I thought wisdom came with age.
400.
by varelse 5y ago
$500K? Hell, ignore net worths below $50M and pile drive into everything above that. But you will then find out just how many sub millionaires erroneously believe they are temporarily embarrassed billionaires who will fight any attempt to r
401.
by varelse 5y ago
Nothing like a barrier to entry to the top 1% to keep out the rabble! But yes, you've discovered the crab pot cheat code that will continue to keep things exactly where they've been and they will remain. Go ahead and propose a 1%
402.
by varelse 5y ago
This is simultaneously everything right and everything wrong with America: an entrepreneurial solution to a problem that shouldn't exist in the first place.
403.
by varelse 5y ago
If you go through the trouble of making your own like Albert Hofmann did, sure. You're not going to though, are you? So you're putting your biggest asset in the hands of criminals of unknown intent. Good luck with that... But I su
404.
by varelse 5y ago
Doubly so if you took the DMT in 2021.
405.
by varelse 5y ago
And wake up in a hospital weeks later realizing you got a bad batch because these drugs aren't legal so there's no quality control and now you're on the fast track to early dementia. Pass.
406.
by varelse 5y ago
Tell that to Justin Zhu. But it's not just LSD and mushrooms. It's nootropics and stimulants too just to keep up in the Red Queen's Race of Silicon Valley at places like AWS. And I know people firsthand who are to up these sh
407.
by varelse 5y ago
If you take career advice from the brainfarts of thought leaders in any other field besides the one you're intent to join, you're going to have a bad time. And even then, thought leaders rarely build, but they sure love the sound
408.
by varelse 6y ago
It is a funny moment. When I die, if people care, may they reminisce on all my quirks and roast me for moments like this. I won't care, I'll be dead.
409.
by varelse 6y ago
I'm pretty sure it's not reactant grade MSG. But I wonder about breakdown products and adulterants that travel with kitchen-grade MSG.
410.
by varelse 6y ago
So I have an allergy to Aloe. And it got blind tested one night when I slept on a friend's floor and woke up with a rash 100% unaware he used his Aloe plant right where I slept. I also have whatever the hell the MSG allergy is, though
411.
by varelse 6y ago
Buy yourself an air quality monitor and experience the wonder of how fast your household air becomes dangerously crappy with the windows closed. The formaldehyde will be with us always apparently...
412.
by varelse 6y ago
AKA you can make whatever you convince someone else to pay you...
413.
by varelse 6y ago
Funny how a presumed engineer lectures scientists about getting political by getting political himself. Aren't you supposed to let societies figure out how to best respond as well? Would you like the authors of this article to do all y
414.
by varelse 7y ago
I am far more excited by the underlying Wafer Scale Integration moonshot than I am by any AI benchmarks here. I know it's trendy to think there can only be one w/r to the AI Iron Throne but nope, not the case, everyone is writing
415.
by varelse 7y ago
Look no further than dredging up Steven Pinker to not only defend "Quantum Supremacy" but to double-down and revisit the NIPS to NEURIPS name change because we apparently have nothing better to talk about on Christmas... https:&#
416.
by varelse 7y ago
So the memory of that film has led me to propose my own absurd DUI sliding scale: If you are caught above at or above that low-end legal limit, you are sentenced to drive your car through an obstacle course filled with clearly marked landmi
417.
by varelse 7y ago
I remember watching an anti-alcohol video in high school wherein they gave 4 or 5 people one drink after another and had them drive an obstacle course increasingly intoxicated. All but one of them got worse with additional alcohol, but the
418.
by varelse 7y ago
High on the technical track as an IC that made multiple significant moneymaking contributions over the past decade (which was used against me last year at my last employer because I hadn't done sufficient pure research in that past dec
419.
by varelse 7y ago
Fully expecting utter disbelief and skepticism, the best information I have indicates the compensation cap at FB is ~$3M. I have personally made up to $1.5M in a year doing AI work. If I had a higher threshold for pedigree BS in tech, I cou
420.
by varelse 7y ago
The downvotes for you are really disappointing. Because before AI, there was the walled garden of HPC, with temple-like supercomputers to which gaining access involved magic incantations and politically savvy groveling from academics wishi

More ›