2 ms·
The language is not really related to this; people have been doing this type of thing in Prolog and Lisp since before the 90s. Actually, with predicates to repr
by anonzzzies 3y ago
The language is not really related to this; people have been doing this type of thing in Prolog and Lisp since before the 90s. Actually, with predicates to represent facts that you learn in your first Prolog class, you can do what you did but in any order, so not just following the hashmap from start to end but create any sentence starting from any point without effort. (Note also there the language doesn’t matter, you can do this in any language, just for Prolog this is the first thing you will learn as it’s a fundament and strength of the language).
I recommend a read of symbolic ai, or better, just a generic ai book: I was raised on Norvig (90s) but no idea how up to date that is kept/still is.
- _akhe 3y agoYep it can work for "next-pixel prediction" too where the token is a pixel. If I have enough samples of arrangements of pixels, it's not fundamentally different than samples of arrangements of words. I'll check out that book - thanks!