HNHacker News
TopNewBestAskShowJobs

adamsmith143

701 karma · joined February 10, 2022

submissionscomments
adamsmith143··on fMRI-to-image with contrastive learning and diffusion priors
>People seem to find "if I don't then someone else will" to be an unacceptable answer but it seems to me to be fairly central.

Because you can use this as a cop out for truly heinous work. I.e. gain of function research, autonomous weapons, chemical weapons, etc. It's not a coherent world view for someone that actually cares about doing good.

adamsmith143··on How to be a consultant, a freelancer, or an independent contractor (2009)
But 32 hours a week is far less than 8-10 hours a day for 6 days. Not sure what you are getting at.
adamsmith143··on It's not mathematics that you need to contribute to (2010)
Just want to point out how cool it is that a Field Medal winner is on MO. Don't know if any Turing Award winners or someone like Jeff Dean is on StackOverflow.
adamsmith143··on Generative AI space and the mental imagery of alien minds
>Then the AI got better, and still the dreamlike things never go away. The constant problem with the hands, words never properly formed, clocks always look wrong, etc.

Might be a bit behind the times, Midjourney seems to have completely solved the hand issue from what I can tell.

adamsmith143··on Bad numbers in the “gzip beats BERT” paper?
Whats disturbing to me if this is true is the number of "top voices" in ML on ML-Twitter were head over heals with this paper. How many of them actually read it at all?
adamsmith143··on An Introduction to Statistical Learning with Applications in Python
People like you are hilarious. You're sitting high in your ivory tower thinking that no one without a PhD in CS from Stanford/Berkeley/MIT can do what you do. Meanwhile people will take the Fast.ai course and be training full llm's from scratch in 6 months all the while you moan that "they don't even understand the REAL math". Yawn.
adamsmith143··on An Introduction to Statistical Learning with Applications in Python
What does that have to do with ML/DL? You don't need Lie Algebra to understand how to do a matrix dot product...
adamsmith143··on Excellence is a habit, but so is failure
> No, the challenge is recognising that food is not equal, and we have chemical and physical reasons why we overeat some and not others.

Those reasons still add up to Calories in - Calories Out. You need Out to be > than In or you will not lose weight. Eat all the kale and celery you want if you aren't burning more than you eat you won't lose weight.

adamsmith143··on Excellence is a habit, but so is failure
They're disputing that literally in this comment thread.
adamsmith143··on Excellence is a habit, but so is failure
Are you really saying that you don't think someone eating 10,000 calories of Kale won't get obese?
adamsmith143··on An Introduction to Statistical Learning with Applications in Python
Interesting seeing job postings wanting 5+ years of LLM experience. Like unless you were at OpenAI working on GPT-1 or Google on BERT there's no one else in the world with that much experience and your shitty Startup/Fortune 500 company can't afford them anyway.
adamsmith143··on An Introduction to Statistical Learning with Applications in Python
This is pure gatekeeping. The math behind LLMs, that is, the math behind Neural Nets, is undergrad freshman level Calculus and some linear algebra. Not really complex at all. Can you deal with derivatives, the chain rule and matrix multiplications? Great you know all the "math" behind Deep Learning.
adamsmith143··on Excellence is a habit, but so is failure
"No one becomes obese eating natural foods" That is the most hippy BS I have ever heard. Eat 10,000 calories of raw sugarcane a day for 1 year and get back to me on the results.

>If you remove the sense of taste from someone, I assure you they won't enjoy food and will lose weight very quickly. Almost like that was the answer all along.

Yes, reducing the number of calories you eat is how you lose weight.

adamsmith143··on Vit D supplementation and major cardiovascular events: D-Health randomised trial
Don't think you need a Phd to understand this: "the confidence interval was consistent with a null finding." Aka the study found nothing.
adamsmith143··on Excellence is a habit, but so is failure
Weird then that Semaglutide is so effective for weight loss when it's mechanism of action is to reduce the amount you eat. Almost like that was the answer all along.
adamsmith143··on Excellence is a habit, but so is failure
Didn't Edison fail like >2000 times when creating the lightbulb? Seems like a "habit" of failure... /s
adamsmith143··on Vit D supplementation and major cardiovascular events: D-Health randomised trial
Can't publish a paper with no findings so they add "might reduce". By the same logic it "Might" increase all cause mortality or might increase lung cancer incidence. Trash study.
adamsmith143··on Responding to “Are bugs and slow delivery ok?”
Maybe, just maybe, the best people to build complex software systems that serve millions of people aren't the ones who can do the most leetcode problems.
adamsmith143··on Who cares about the Ivy League?
It happens much more in Asia, a place with oh idk several billion people. So hardly a US phenomena.
adamsmith143··on Who cares about the Ivy League?
> I'll bet that most students with good grades who are applying to elite colleges have higher than average IQs.

That wasn't your claim:

> You're assuming that people who are admitted to Ivy League schools a) have higher IQs than average and b) have parents with higher IQs than average.

>What differentiates the ones who get accepted from the ones who don't? At Caltech and MIT?

Possibly the most meritocratic schools in the US?

Idk, .001 on your GPA or a single point on your SAT and AP scores? Or the fact that you didn't win an IOM medal or Intel Talent Search?

adamsmith143··on AI weights are not open “source”
The question shouldn't be whether the weights are copyrightable but whether they are protected under other electronic communication/data privacy laws.
adamsmith143··on AI weights are not open “source”
Corporate data is of course protect-able. Otherwise why don't you just open up all your databases so anyone can access them?
adamsmith143··on Who cares about the Ivy League?
>No, it's a cultural thing.

IIT's, Peking/Tsinghua, SNU(Korea), University of Tokyo, Oxbridge, etc. etc. Cultural thing that happens across the entire world?

adamsmith143··on Who cares about the Ivy League?
Are you saying that say MIT and Caltech students don't have higher than average IQ? It's almost a certainty that they do and I'm sure it extends to the Ivy league and other elite schools.
adamsmith143··on US Congress doubles down on claims of illegal UFO retrieval programs
I'm talking about the constituents who actually believe that the reason not to adopt nationalized healthcare is because the gov would screw it up.
adamsmith143··on US Congress doubles down on claims of illegal UFO retrieval programs
You mean the previous Prez's son in law who somehow got a Multibillion dollar investment from the Saudi's while he ostensibly worked directly in the white house? Yeah that should be investigated for sure.
adamsmith143··on US Congress doubles down on claims of illegal UFO retrieval programs
Right they simultaneously believe the government is so incompetent they can't implement national healthcare that every other major nation on earth has figured out but are so competent they can get thousands of people to keep the most impactful secret in the history of mankind.
adamsmith143··on Profits are driving inflation. When will politicians and the BoE admit it?
Sorry to burst your bubble but the real market doesn't work like the ones you learn about in Econ 101.
adamsmith143··on Bootstrapping Atlassian: $10k In Credit Card Debt To $20M In Revenue In 5 years
Classic survivorship bias, 100s of other startups went bankrupt and left their founders with massive high interest debt.
adamsmith143··on New York State Senate passes prohibitions on non-competes
Hedge Funds and I-Banks are on suicide watch.
← PreviousPage 2 of 31Next →