AI engineering glossary · Calculus optimisation
What is JVP (jacobian-vector product)?
Jacobian-vector product (also called JVP): The product of a Jacobian matrix with a vector, giving how all outputs change when the inputs move in one chosen direction. Forward-mode differentiation computes it without forming the Jacobian.
Jacobian-vector product: an example
In JAX, jax.jvp(f, (x,), (v,)) returns f(x) together with J(x) times v for a direction v.
Related terms
More Calculus optimisation terms
- KKT conditions
- L-BFGS
- L-smoothness
- Lagrange multiplier
- Lagrangian
- Learning rate
- Learning rate schedule
- Learning rate warmup
Where you learn it
Jacobian-vector product is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.