valegrete

4 ms·

1. 1.▲
   by valegrete 3d ago

   How do you do that in the current paradigm other than creating yet another gameable metric? And something I didn't mention above is that there is no difference between "solving the task" and "optimizing the metric"
2. 2.▲
   by valegrete 3d ago

   This is the problem with optimization generally, even in the human domain. You measure task performance with a metric and punish/reward based on the metric. Anyone who likes reward / hates punishment isn't going to actually c
3. 3.▲
   by valegrete 4d ago

   I just wonder whether a lot of smart people who never needed to go beyond the "solve for X" algorithmic math of a typical calculus sequence are actually reading the declaration the way the signatories wrote it. The Navier-Stokes p
4. 4.▲
   by valegrete 5d ago

   Mathematical breakthroughs with commercial relevance are few and far between, and often depend on dusting off old results which were, at the time of discovery, "solutions nobody asked for." The NS counterxample is actually, by any
5. 5.▲
   by valegrete 6d ago

   Other people are allowed to have their priors, too. Even under a non-informative prior, the weight of the evidence (Tristan's account, OpenAI's announcement, and Bubeck's "denial", if you want to call it that, plus