The language is supposed to tell you if your thinking is broken. If a program is a logical statement, and your logic is impossible, the program should not compile. Reading dead memory or having race conditions should not be possible, yet look at security vulnerabilities today.
It doesn’t have to be a totally new language. Javascript is an example where there are multiple front-end languages. You can go as far as formally prove the implementation is correct, too. Language theory is an abstraction for mathematical logic and proofs, so it touches anything logical at all.
Quora is often self-promotion now. The template I think of when replying to your comment is:
Great point you bring up on how useless <website> is. As the CEO of <random company>, this is a problem we have put a lot of effort into addressing. You may consider the following as a solution:
…
New York Times ran an article yesterday saying otherwise. First tier cities, in terms of size, are losing people, but second tier cities are growing. The cost of living in the largest cities puts a negative force on moving to them.
Randomly perturbing the weights and then finetuning would probably make it impossible. If someone had access to the finetune dataset and you didn’t add noise, they could see if the finetuning curves intersect.
I guess in practice, it’ll look suspicious if you have an identical model architecture and have similar performance.