What is it?
Continuity where one works for every point at once. On a closed bounded interval every continuous function is uniformly continuous (Heine–Cantor), which makes approximation results such as Weierstrass's theorem possible.
Formulas
Where it shows up in AI
Proofs of the universal approximation theorem use uniform continuity on compact sets.
Where is it used?
Computing topics reachable from here, through the chain of ideas that leads to them:
ℒ AI and machine learning
- Lipschitz continuity→Gradient descent★★★★★
- Lipschitz continuity→Gradient descent→Learning rate★★★★★
- Lipschitz continuity→Gradient descent→Backpropagation★★★★★
- Lipschitz continuity→Gradient descent→Stochastic gradient descent (SGD)★★★★★
- Lipschitz continuity→Gradient descent→Loss landscape★★★★★
- Lipschitz continuity→Gradient descent→Reinforcement learning★★★★★
- +7
⚛ Physics and simulation
- Lipschitz continuity→Initial value problems: existence and uniqueness→Physics engines★★★★★
- Lipschitz continuity→Initial value problems: existence and uniqueness→Physics engines→N-body gravitational simulation★★★★★
- Lipschitz continuity→Initial value problems: existence and uniqueness→Physics engines→Fluid dynamics and CFD★★★★★
- Lipschitz continuity→Initial value problems: existence and uniqueness→Physics engines→Fluid dynamics and CFD→Weather and climate modelling★★★★★
What depends on it
This page has the essentials. A fuller treatment (intuition, formal definition, worked example) is on the way.