Once you frame it as a smooth problem (differentiable), it can be optimized and use hill climbing.

2024-08-19 · Bits and Bobs 8/19/24

One of the reasons that identifying a good "self steering metric" can be powerful.

For example, for rolling out an ambitious technology: "Maximize absolute amount of usage while minimizing the number of users who have such a bad time they'll never use it again."

More on this topic

From other episodes

If it's a hill climbing exercise, first, make sure you're on the right hill.2026-01-06
Someone's personality could emerge even from very small starting biases.2025-11-04
Hill climbing a moving hill doesn't work.2025-04-28
Learning happens in feedback loops.2024-12-09
If you peek into how multi-modal models work, they're a ball of cheap, random hacks that turn out to be terrifically, unreasonably effective.2024-11-11
A judo move: if you can frame a problem as an optimization problem (without dangerous externalities) you can now use hill climbing techniques on it.2024-10-28
LLMs make the "good enough" viable zone larger.2024-10-14
An interesting pattern: using LLMs to astroturf content in an ecosystem.2024-07-01