The concept of probably approximately correct (PAC) learning has been foundational in computational learning theory.
The best thing about being a writer is that you get to live in the world you create.
Every failed experiment is a step closer to success, as long as you learn from it.