3 ms·
I'm not claiming any arbitrary set of weights is a minimal representation. But typically, if people could achieve the same quality of results with a smaller s
by rhelz 27d ago
I'm not claiming any arbitrary set of weights is a minimal representation. But typically, if people could achieve the same quality of results with a smaller set of weights, or weights which have been quantized to lower bit representations, etc, they would have published the smaller one instead.
- andrewflnr 27d agoYou kind of are claiming they're minimal, though. Because if they're not, your statement that "if you found any patterns in there, you could exploit the regularity..." implies nothing. Yeah, the patterns are there, and people are exploiting them. Your socioeconomic argument just doesn't hold either. People don't delay releasing models until they've minimized it to the theoretical limit. They ship it when it's good enough for whatever job they're making it for.
- rhelz 26d agoIf somebody finds patterns in them, that means they can be improved by reducing the patterns. Patterns and symmetry are forms of redundancy. The less memory the weights take to meet a level of competency, the more random, and therefore patternless and inscrutable they will be.
- andrewflnr 26d agoNo kidding. Clearly there's some redundancy, or however you want to frame it, in real-world released models. That still means your argument for why they can never be interpretable is based an incorrect premise.