In the pre-training set, there are lots of places where humans talk about causality (both informally and more formally in myriad academic papers). So a model would ultimately need to learn abstract stuff about causality (e.g. correlation is not causation, arrow of time, causes are local, etc) and concrete causal facts (the moon causes tides, tiny organisms cause mold, etc). Given this knowledge, it’s plausible a model M could make reasonable guesses for questions like, “What happens when a model with [properties of model M] starts interacting with the world?” These guesses would be improved by finetuning by RL on actual interaction between M and the world.
(It seems that most of what my ability to make OOD predictions or causal inferences is based on passive/offline learning. I know science from books/papers and not from running my own rigorous control experiments or RCTs.)
I disagree with your last point. Since we’re agents, we can get a much better intuitive understanding of what causality is, how it works and how to apply it in our childhood. As babies, we start doing lots and lots of experiments. Those are not exactly randomized controlled trials, so they will not fully remove confounders, but it gets close when we try to do something different in a relatively similar situation. Doing lots of gymnastics, dropping stuff, testing the parent’s limits etc., is what allows us to quickly learn causality.
LLMs, as they are currently trained, don’t have this privilege of experimentation. Also, LLMs are missing so many potential confounders as they can only look at text, which is why I think that systems like Flamingo and Gato are important (even though the latter was a bit disappointing).
I agree my last point is more speculative. The question is whether vast amounts of pre-trained data + a smaller amount of finetuning by online RL substitutes for the human experience. Given the success of pre-training so far, I think it probably will.
Note that the modern understanding of causality in stats/analytic philosophy/Pearl took centuries of intellectual progress—even if it seems straightforward. Spurious causal inference seems ubiquitous among humans unless they have learned—by reading/explicit training—about the modern understanding. Your examples from human childhood (dropping stuff) seem most relevant to basic physics experiments and less to stochastic relationships between 3 or more variables.
I can interpret your argument as being only about the behavior of the system, in which case: - I agree that models are likely to learn to imitate human dialogue about causality, and this will require some amount of some form of causal reasoning. - I’m somewhat skeptical that models will actually be able to robustly learn these kinds of abstractions with a reasonable amount of scaling, but it certainly seems highly plausible.
I can also interpret your argument as being about the internal reasoning of the system, in which case: - I put this in the “deep learning is magic” bucket of arguments; it’s much better articulated than what we said though, I think... - I am quite skeptical of these arguments, but still find them plausible. I think it would be fascinating to see some proof of concept for this sort of thing (basically addressing the question ‘when can/do foundation models internalize explicitly stated knowledge’)
I’m somewhat skeptical that models will actually be able to robustly learn these kinds of abstractions with a reasonable amount of scaling
GPT-3 (without external calculators) can do very well on math word problems (https://arxiv.org/abs/2206.02336) that combine basic facts about the world with abstract math reasoning. Why think that the kind of causal reasoning humans do is out of reach of scaling (especially if you allow external calculators)? It doesn’t seem different in kind from these math word problems.
when can/do foundation models internalize explicitly stated knowledge
Some human causal reasoning is explicit. Humans can’t do complex and exact calculations using System 1 intuition, and neither can we do causal reasoning of any sophistication using System 1. The prior over causal relations (e.g. that without looking at any data ‘smoking causes cancer’ is way more likely than the reverse) is more about general world-model building, and maybe there’s more uncertainty about how well scaling learns that.
RE GPT-3, etc. doing well on math problems: the key word in my response was “robustly”. I think there is a big qualitative difference between “doing a good job on a certain distribution of math problems” and “doing math (robustly)”. This could be obscured by the fact that people also make mathematical errors sometimes, but I think the type of errors is importantly different from those made by DNNs.
This is a distribution of math problems GPT-3 wasn’t finetuned on. Yet it’s able to few-shot generalize and perform well. This is an amazing level of robustness relative to 2018 deep learning systems. I don’t see why scaling and access to external tools (e.g. to perform long calculations) wouldn’t produce the kind of robustness you have in mind.
I think you’re moving the goal-posts, since before you mentioned “without external calculators”. I think external tools are likely to be critical to doing this, and I’m much more optimistic about that path to doing this kind of robust generalization. I don’t think that necessarily addresses concerns about how the system reasons internally, though, which still seems likely to be critical for alignment.
The important part of his argument is in the second paragraph, and I agree because by and large, pretty much everything we know about science and casuality, at least in the beginning for AI is on trusting the scientific papers and experts. Virtually no knowledge is given by experimentation, but instead by trusting the papers, experts and books.
That might be a crux here, since I view a lot of our knowledge of causality and physics essentially we take on trust, so that we don’t need to repeat experimentation.
In the pre-training set, there are lots of places where humans talk about causality (both informally and more formally in myriad academic papers). So a model would ultimately need to learn abstract stuff about causality (e.g. correlation is not causation, arrow of time, causes are local, etc) and concrete causal facts (the moon causes tides, tiny organisms cause mold, etc). Given this knowledge, it’s plausible a model M could make reasonable guesses for questions like, “What happens when a model with [properties of model M] starts interacting with the world?” These guesses would be improved by finetuning by RL on actual interaction between M and the world.
(It seems that most of what my ability to make OOD predictions or causal inferences is based on passive/offline learning. I know science from books/papers and not from running my own rigorous control experiments or RCTs.)
I disagree with your last point. Since we’re agents, we can get a much better intuitive understanding of what causality is, how it works and how to apply it in our childhood. As babies, we start doing lots and lots of experiments. Those are not exactly randomized controlled trials, so they will not fully remove confounders, but it gets close when we try to do something different in a relatively similar situation. Doing lots of gymnastics, dropping stuff, testing the parent’s limits etc., is what allows us to quickly learn causality.
LLMs, as they are currently trained, don’t have this privilege of experimentation. Also, LLMs are missing so many potential confounders as they can only look at text, which is why I think that systems like Flamingo and Gato are important (even though the latter was a bit disappointing).
I agree my last point is more speculative. The question is whether vast amounts of pre-trained data + a smaller amount of finetuning by online RL substitutes for the human experience. Given the success of pre-training so far, I think it probably will.
Note that the modern understanding of causality in stats/analytic philosophy/Pearl took centuries of intellectual progress—even if it seems straightforward. Spurious causal inference seems ubiquitous among humans unless they have learned—by reading/explicit training—about the modern understanding. Your examples from human childhood (dropping stuff) seem most relevant to basic physics experiments and less to stochastic relationships between 3 or more variables.
Well maybe llms can “experiment” on their dataset by assuming something about it and then being modified if they encounter counterexample.
I think it vaguely counts as experimenting.
I can interpret your argument as being only about the behavior of the system, in which case:
- I agree that models are likely to learn to imitate human dialogue about causality, and this will require some amount of some form of causal reasoning.
- I’m somewhat skeptical that models will actually be able to robustly learn these kinds of abstractions with a reasonable amount of scaling, but it certainly seems highly plausible.
I can also interpret your argument as being about the internal reasoning of the system, in which case:
- I put this in the “deep learning is magic” bucket of arguments; it’s much better articulated than what we said though, I think...
- I am quite skeptical of these arguments, but still find them plausible. I think it would be fascinating to see some proof of concept for this sort of thing (basically addressing the question ‘when can/do foundation models internalize explicitly stated knowledge’)
GPT-3 (without external calculators) can do very well on math word problems (https://arxiv.org/abs/2206.02336) that combine basic facts about the world with abstract math reasoning. Why think that the kind of causal reasoning humans do is out of reach of scaling (especially if you allow external calculators)? It doesn’t seem different in kind from these math word problems.
Some human causal reasoning is explicit. Humans can’t do complex and exact calculations using System 1 intuition, and neither can we do causal reasoning of any sophistication using System 1. The prior over causal relations (e.g. that without looking at any data ‘smoking causes cancer’ is way more likely than the reverse) is more about general world-model building, and maybe there’s more uncertainty about how well scaling learns that.
RE GPT-3, etc. doing well on math problems: the key word in my response was “robustly”. I think there is a big qualitative difference between “doing a good job on a certain distribution of math problems” and “doing math (robustly)”. This could be obscured by the fact that people also make mathematical errors sometimes, but I think the type of errors is importantly different from those made by DNNs.
This is a distribution of math problems GPT-3 wasn’t finetuned on. Yet it’s able to few-shot generalize and perform well. This is an amazing level of robustness relative to 2018 deep learning systems. I don’t see why scaling and access to external tools (e.g. to perform long calculations) wouldn’t produce the kind of robustness you have in mind.
I think you’re moving the goal-posts, since before you mentioned “without external calculators”. I think external tools are likely to be critical to doing this, and I’m much more optimistic about that path to doing this kind of robust generalization. I don’t think that necessarily addresses concerns about how the system reasons internally, though, which still seems likely to be critical for alignment.
The important part of his argument is in the second paragraph, and I agree because by and large, pretty much everything we know about science and casuality, at least in the beginning for AI is on trusting the scientific papers and experts. Virtually no knowledge is given by experimentation, but instead by trusting the papers, experts and books.
I disagree; I think we have intuitive theories of causality (like intuitive physics) that are very helpful for human learning and intelligence.
That might be a crux here, since I view a lot of our knowledge of causality and physics essentially we take on trust, so that we don’t need to repeat experimentation.