Survivor Bias

What people said

sillysaurus3 · 2018-08 · Strong recommendation

Happy to see Veritasium getting some attention on HN. You may have seen his previous video from 2014, Facebook Fraud: [LINK] One of my favorite videos is on his alternate channel, 2veritasum. Survivor Bias: [VIDEO]

Read the original HN comment ↗

stevenjgarner · 2022-06 · Reference

IMHO, this was not cheating. This is a phenomenon called success. If you are fortunate (?) to be around in the entrepreneurial world long enough, you will realize that intelligence, hard work, good time management etc are not adequate. They just qualify as deserving. There are serendipities that separate those who succeed from everybody else who was deserving. Survivor bias [1]. Calling Eli was your serendipity. The opposite is also true: killer asteroids hit the planets of otherwise deserving individuals and initiatives. I was co-founder of a municipal fiber-to-the-home initiative in 1996. We had both a winning technical recipe and also a financial engineering recipe. (Extremely rare even today) We were then fortunate enough to acquire a customer that was a newly minted public company, representing a whopping 65% of our business. Banker's dream, but in actuality a big mistake. Then the double-whammy of fraud by that public company (wiping it out completely before our contract was realized) AND the dot-com crash. Killer asteroids are real. [1] [VIDEO]

Read the original HN comment ↗

sillysaurusx · 2021-05 · Reference

is that actually the case? [are model weights being published less frequently?] It sure feels like it. But it could be survivorship bias: [VIDEO] or regression to the mean: [LINK] . It's possible that StyleGAN and GPT-2 just happened to coincide, and both companies just happened to decide to release their models rather than hoard them, and that the usual case is for companies to rarely (if ever) release weights with their research. But the more interesting question is, why wouldn't they? So onward to your second one: are ethical concerns actually the reasons some models/weights are not released? Or is it about hoarding a potentially valuable commodity? Or some other reason, or a combination of several? As with most things, the answer is likely the mundane, dull truth that big companies tend to have lots of inertia and friction preventing such things from happening. Sources of friction seem to include (but are not limited to): - are we legally responsible for what other people do with the model? (no.) - should we restrict people from using the model commercially? (why bother? but sure, nVidia did that for stylegan. It didn't stop Artbreeder from blatantly ignoring and monetizing it anyway. And while that sounds negative, Artbreeder is one of the coolest ML projects ever, as far as I'm concerned, so the world is better off for ignoring your stupid policy.) - What if people start misusing the model? (So what? I feel like I should just get in the face of whoever is asking this qu

Read the original HN comment ↗