niplav @niplav@schelling.pt

Post: Precision of Sets of Forecasts

niplav.site/precision

I conclude that methods based on rounding probabilities are hot flaming garbage, methods based on rounding odds or log-odds are regular garbage, and methods based on applying noise to log-oddsare ok but sort of slow.

Midnight is the best time to fire off a bunch of aversive emails

Back from 𝕏

The good thing about niplav is that he just isn't that smart

Europe is performing autoerotic asphyxiation on a civilizational level

Today I learned that some people use ℘ to represent the powerset.

Today was a good day, but that was a mediocre and disappointing move on humanity's part

"Space-Value Taxation under the Kerr-Newman Metric"

Positive update on effective altruists: I thought they wouldn't be able to update from technical AI alignment as a most important cause to a new cause X. It seems like that cause X is AI governance and the vibe shifted towards it.
Congrats! You've exceeded my expectations.

Oh cool we doing surreal consequentialism now?

Supervolution

I haven't thouhht about the game theory of it though

Alternative view: holidays just like dropout and makes orgs more robust

It is possible to get more detailed information on employee performance by looking at semi-random holidays:

If their job is still ~getting done (or the department output isn't suffering much), then they're either not working very much or their job can be taken over by coworkers

Buying 4 Nvidia shares two years ago, and 2 more a year ago, was a really good move on my part

The more famous the person the worse the questions