Viewing profile — marbletimes

They are not slightly different, they measure something (totally) different. Confidence intervals are relative to E[y|x], Predictive intervals are relative to y. Sometimes, for exa…

I agree that statisticians would better than CS people appreciate the importance of uncertainty intervals--it is mostly cultural--but that "In reality there are very few ML applica…

From a good SO answer, see https://stats.stackexchange.com/questions/16493/difference-b... "A confidence interval gives a range for E[y∣x], as you say. A prediction interval gives …

Correct, quantile regression is an option. Another is "pure" bootstrapping (you can see by googling something like uncertainty + machine learning + bootstrapping that this is a ver…

When I was in academia, I used to fit highly sophisticated models (think many-parameters, multi-level non-linear mixed effect models) who were giving not only point estimate but al…

But if you know that the current company cannot pay like the other company you are interviewing for, or that in this new company the future salary outlook is much better, why bothe…

The assumption here is that the employee leaving the company was due to some problems that could not be solved after careful discussion with the manager, skip-level etc. But leavin…

Jiu-Jitsu 3 classes a week for 1.5 hours each. Weight training, 3 times a week, 1 hour each time. Running, 2 times a week, 45 minutes each run. That's on average, total 9 hours a w…

"My theory is that he’s too valuable where he is, and that they don’t want to move him elsewhere." Assuming the "they are too valuable here to be promoted elsewhere" hypothesis is …

The hypothesis here is that there is a trade-off between making bank and living a pleasant life, whatever attributes one wants to associate with "pleasant", maybe low professional …

Agreed. That's why I wrote,"but I consider that money far from ----financially---- making someone "a very successful 40-year-old"."

As someone living in the United States and earning much more than I would in my home country, even though my professional contribution would be the same if the same opportunities e…

It depends on where. In a European country, think Spain, Italy, France, "get good grades, study a valuable major, don't eat garbage, don't marry crazy people/idiots, stay off dange…

What makes the real difference is curating followers and avoiding mindless "reel type" scrolling. As always, it's being intentional rather than setting hard limits (e.g., being fri…

Feynman certainly cared a great deal about what others thought of him, and he has been known to embellish many stories to appear as the common genius in the situation. In short, he…

Let's not forget the lies that are broadcasted left and right on social media, in which thousands become millions, and trite tropes are sold like congress-worthy insights. You are …

It is a tricky subject. When I referred to a model I had published elsewhere, I was asked to write about the model again in the current paper, since readers may not be able to acce…

When I was in academia--before I moved into technology and began applying statistical modeling and machine learning to a variety of tech problems--my research in mathematical biolo…

This is one of those witty sentences that sound good (the balance of life, laughs here, tears there, if you want to enjoy living you need to accept dying, everything happens for a …

It is one of those common tropes, "interviewing is hard", that seems to justify the inevitability of poor attitude, wasting time of interviewees (remember the interviewers are paid…