3 ms·

That's true. I don't think humans are very good at general intelligence. Our genes give us priors to evade lions and pick berries. Humanity as cultural organi

That's true. I don't think humans are very good at general intelligence. Our genes give us priors to evade lions and pick berries. Humanity as cultural organism can reprogram us to more abstract tasks, but it takes easily 10-20 years of training and we are relatively inefficient in them compared to our original task. But even then. I don't think current in-distribution learning is enough even if we scale it up. There must be something else. We are at least 3-5 Turing awards away from AGI, I think.

jfengel 6y ago
Probably. But we were more like 10 to 20 Turing awards way just a decade ago. I don't expect progress to be consistent; those remaining awards could take another century. But we made what might be a huge amount of progress in a very short time. Or it could be a complete dead end, as has happened before. And nobody could tell until it was done; even those who were correctly skeptical were really just lucky that there wasn't a rabbit in that hat. Sometimes there's a rabbit.
phreeza 6y ago
In the general sense, is out-of-distribution learning even possible? The best you are likely going to get is robustness wrt different environments, but the environments have to come from the same Meta-Distribution. Invariant risk minimization is an interesting research area in that direction, there is for example an adversarial formulation that seems quite elegant. I have never tried this, but I can imagine it is tricky to get to converge.
- MAXPOOL 6y ago
  >In the general sense, is out-of-distribution learning even possible? Yes. Humans can learn to answer questions even when questions (test cases) don't come from the same distribution as the training set. You need to have some amount of analytical capabilities. Analytic as the ability to form distinct and modular concepts from the input and form representations that can be composed together to generate instances that don't exist in the distribution. The low-level example of this ability could be blind source separation.
  - phreeza 6y ago
    I guess it is a question of definitions, but i think it really is not possible, for the reason that the space of possible test sets includes contradictory sets. So if I ask you a simple question which is not in your training set, like "Does a Quigloth have eyes?", there can be two possible test sets, one where the answer is no, and one where the answer is yes. So a reply to this could be that questions have to be about the real world. But that is then the same "meta distribution" as I mentioned in my previous comment, and not out of distribution learning in the pure sense.

jfengel 6y ago

Probably. But we were more like 10 to 20 Turing awards way just a decade ago. I don't expect progress to be consistent; those remaining awards could take another century. But we made what might be a huge amount of progress in a very short time. Or it could be a complete dead end, as has happened before. And nobody could tell until it was done; even those who were correctly skeptical were really just lucky that there wasn't a rabbit in that hat. Sometimes there's a rabbit.

phreeza 6y ago

In the general sense, is out-of-distribution learning even possible? The best you are likely going to get is robustness wrt different environments, but the environments have to come from the same Meta-Distribution. Invariant risk minimization is an interesting research area in that direction, there is for example an adversarial formulation that seems quite elegant. I have never tried this, but I can imagine it is tricky to get to converge.

MAXPOOL 6y ago
>In the general sense, is out-of-distribution learning even possible? Yes. Humans can learn to answer questions even when questions (test cases) don't come from the same distribution as the training set. You need to have some amount of analytical capabilities. Analytic as the ability to form distinct and modular concepts from the input and form representations that can be composed together to generate instances that don't exist in the distribution. The low-level example of this ability could be blind source separation.
- phreeza 6y ago
  I guess it is a question of definitions, but i think it really is not possible, for the reason that the space of possible test sets includes contradictory sets. So if I ask you a simple question which is not in your training set, like "Does a Quigloth have eyes?", there can be two possible test sets, one where the answer is no, and one where the answer is yes. So a reply to this could be that questions have to be about the real world. But that is then the same "meta distribution" as I mentioned in my previous comment, and not out of distribution learning in the pure sense.