auraai

Their intelligence was clear to me growing up around them, without knowing anything about crows specifically. They clearly have intelligent interactions with each other and the environment, and have a "theory of mind" of people approaching them/interacting with them. They could also figure out how to open/untie garbage bags, gang up to bully other birds and take their food, etc. In the mornings, right outside our house, they would congregate on a tall pine tree and it really looked like they were having a daily standup - one of them yells loudly, others listen silently, then start making noise together :)

A model's need for data is a sort of reciprocal to its inductive bias strength. The more permissive is your model (can learn anything/fit noise perfectly), the more data you need to tune it to a useful state. Conversely, the more restrictive is your model (e.g. y = ax + b), the less data you need (e.g. two points).

People needed a lot of data to predict the movement of planets (entire books of numeric tables), until laws of gravity were figured out, at which point it was reduced to a couple of parameters. This same principle applies to modern AI too, the more you restrict your inductive bias to the sort of structures and dynamics you expect to capture in the wild, the less volumes of data you need to tune.

So is "data a bad idea"? Only as bad as your world model is good. Perfect model of the world requires zero data, weak model of the world requires lots of data.

There's lots of room for cross-pollination between bio/life sciences and ML/AI. One key insight is the importance of what you pick as your primary representation of data (is everything a number, a symbolic structure, a probability distribution, etc). I believe a lot of these bio-inspired approaches over-emphasize the embodied nature of intelligence and how much it needs to be situated in space and time, which downplays all the sub-problems that need to be solved in other "spaces" with less obvious "spatiotemporal" structure. I believe space and time are emergent, at least for the purposes of defining intelligence, and there are representations where both space and time arise as dimensions of their structure and evolution.

This is pretty important in mathematical finance, where one moves from a real-world measure to a risk-neutral measure to make computations feasible.

https://en.wikipedia.org/wiki/Girsanov_theorem https://en.wikipedia.org/wiki/Risk-neutral_measure

Poincaré once said that math isn't about numbers, but relationships between numbers

He described math more poetically and profoundly: "the art of giving the same name to different things"

We seem to be crawling out of the "uncanny valley" on the other side. Give it a couple more years, we've pretty much nailed the interpolation of human output (across conventional modalities).

I don't think this type of method can scale indefinitely, it's essentially just "better" sampling within dense areas of knowledge space. It cannot help with better exploration outside these dense areas, because these explorations won't have a consensus among agents almost by definition.