(Half jokingly) What about using imitation learning followed by reinforcement learning? In other words, start by looking at what kinds of posts your role models write, try to infer when they tend to start writing, do the same thing and then make adjustments as you go using RL. I think this is what I did (subconsciously).
If you want a more explicit kind of answer, I think I need a better idea of what your dilemma is. You wrote:
If I explain my ideas now, I’m going to be embarrassed by it next year.
but also:
This worry is not about impressions.
which confuses me because it seems like worrying about being embarrassed is worrying about impressions?
which confuses me because it seems like worrying about being embarrassed is worrying about impressions?
What I meant to say is that I can tell that my work isn’t going to be very good from next year’s standards, which are better standards because they’re more informed
Why is that a problem, unless you’re worried about wasting people’s time with half-baked ideas (which you said that you’re not)? Surely we should just judge our work by our current standards, otherwise we might as well judge our work by what we expect post-Singularity standards to be and then never publish anything?
Because some people might already be at this level, and I worry that I’m just adding noise to their signal.
Maybe my question is this: given that, every year, I unexpectedly learn important considerations that discredit my old beliefs, how can I tell that my models are further along this process than those written by others?
I guess you can either write down your current models and ask people privately for feedback, or just talk to people who you think might have better models than you and try to learn from them. Write down your models for public consumption when your private feedback/learning suggests that your models are as good as the state of the art, or at least competitive with what has been publicly written down.
Also I’m curious, what important considerations have you learned recently?
I think if you write, next year’s standards will be even better standards because you’ll improve. And if you want, you’ll be able to write them better, and if not, stuff you write after that will be better.
(Half jokingly) What about using imitation learning followed by reinforcement learning? In other words, start by looking at what kinds of posts your role models write, try to infer when they tend to start writing, do the same thing and then make adjustments as you go using RL. I think this is what I did (subconsciously).
If you want a more explicit kind of answer, I think I need a better idea of what your dilemma is. You wrote:
but also:
which confuses me because it seems like worrying about being embarrassed is worrying about impressions?
What I meant to say is that I can tell that my work isn’t going to be very good from next year’s standards, which are better standards because they’re more informed
Why is that a problem, unless you’re worried about wasting people’s time with half-baked ideas (which you said that you’re not)? Surely we should just judge our work by our current standards, otherwise we might as well judge our work by what we expect post-Singularity standards to be and then never publish anything?
Because some people might already be at this level, and I worry that I’m just adding noise to their signal.
Maybe my question is this: given that, every year, I unexpectedly learn important considerations that discredit my old beliefs, how can I tell that my models are further along this process than those written by others?
I guess you can either write down your current models and ask people privately for feedback, or just talk to people who you think might have better models than you and try to learn from them. Write down your models for public consumption when your private feedback/learning suggests that your models are as good as the state of the art, or at least competitive with what has been publicly written down.
Also I’m curious, what important considerations have you learned recently?
I think if you write, next year’s standards will be even better standards because you’ll improve. And if you want, you’ll be able to write them better, and if not, stuff you write after that will be better.