lukeor

5 ms·

1.
by lukeor 5y ago
I never said radiology didn't produce biased results, just that we don't know the race of our patients most of the time. There are lots of ways bias can still occur, like in who gets referred for scans, when they get referred, wha
2.
by lukeor 5y ago
Hi, author here. I'll try to answer here with two points. 1) race is not provided explicitly or intentionally during training, but medical practice is biased so it is reasonable to assume there is a signal in the data. 2) we know there
3.
by lukeor 7y ago
I don't actually think imagenet is anywhere near as susceptible to crowd based overfitting as most kaggle competitions, but I don't actually think that paper falsifies the claim that it is. That paper shows that imagenet classifie
4.
by lukeor 7y ago
All I can really say is that my usual readers understand that I am pro-ML, in fact I'm probably more hung go about the potential of deep learning than many of my compatriots. I've fallen victim of getting a Twitter bump, and assum
5.
by lukeor 7y ago
Hi, author here. There are a range of ways the estimates can be improved, although many require data that isn't available. The main point is that having a ballpark idea of how reliable your results are is good, and you can achieve that
6.
by lukeor 7y ago
Hi, author here. I didn't actually mean to suggest that the last 5 years of performance improvement could be spurious. That clearly isn't true. I use resnets/densenets etc in my day to day work! What the picture was trying to
7.
by lukeor 9y ago
Hey, thanks for liking the article. I was just looking at my blog stats and they had a bit of spike when you wrote this. I'm just getting down to writing some posts about the big question the New Yorker article introduces, but doesn&#x
8.
by lukeor 10y ago
Hi, I'm the author of the blogpost (didn't expect to see it here!) and I largely agree with you. I think I would say that the use of private datasets isn't actually a problem though, the problem is that the methods and analys