Why can a solution exist but not be unique?

Existence needs only continuity of the equation's right-hand side, but uniqueness requires that the right-hand side not change too steeply, typically a Lipschitz condition. The equation y' equal to the square root of the absolute value of y, with zero initial value, is a standard example admitting more than one solution.

What does the Picard iteration actually do?

It rewrites the initial value problem as an integral equation and repeatedly substitutes an approximate solution into the integral. When the right-hand side is Lipschitz, this iteration is a contraction, so it converges to the unique fixed point, which is the sought solution.

Existence and Uniqueness Theorems

Existence and uniqueness theorems state the conditions under which an initial value problem for an ordinary differential equation has a solution and exactly one solution.

Definition

An existence theorem asserts that a solution to an initial value problem exists on some interval; a uniqueness theorem asserts that, under stronger hypotheses such as a Lipschitz condition on the right-hand side, no two distinct solutions can share the same initial value.

Scope

This topic covers the Picard-Lindelof theorem and its proof by successive approximations and the contraction mapping principle, Peano's existence theorem under mere continuity, Gronwall's inequality and continuous dependence on initial data, and the continuation of solutions and maximal intervals of existence.

Core questions

Under what conditions does an initial value problem possess a solution?
What additional hypothesis guarantees that the solution is unique?
How far in time can a solution be continued before it ceases to exist?
How sensitively does the solution depend on its initial data?

Key theories

Picard-Lindelof theorem: If the right-hand side is continuous and Lipschitz in the dependent variable, the initial value problem has a unique solution on a neighborhood of the initial point, obtained as the limit of Picard iterates via the contraction mapping principle.
Peano existence theorem: Continuity of the right-hand side alone guarantees existence of at least one solution, but without a Lipschitz condition uniqueness can fail, as classic examples with non-unique solutions show.
Gronwall inequality and continuous dependence: Gronwall's inequality bounds a function satisfying an integral inequality, and it yields uniqueness and the continuous dependence of solutions on initial conditions and parameters.

Clinical relevance

These theorems justify treating a model's solution as a well-defined object: they tell modelers when a differential equation determines a unique trajectory from given data, a prerequisite for prediction, numerical simulation, and the qualitative theory of dynamical systems.

History

Cauchy gave the first existence proofs in the 1820s, and Lipschitz isolated the condition now bearing his name. Picard's method of successive approximations and Lindelof's contributions yielded the constructive theorem standard today, while Peano showed in 1886 that continuity alone secures existence though not uniqueness.

Key figures

Augustin-Louis Cauchy
Rudolf Lipschitz
Emile Picard
Ernst Lindelof
Giuseppe Peano

Seminal works

coddington1955
hartman2002

Frequently asked questions

Why can a solution exist but not be unique?: Existence needs only continuity of the equation's right-hand side, but uniqueness requires that the right-hand side not change too steeply, typically a Lipschitz condition. The equation y' equal to the square root of the absolute value of y, with zero initial value, is a standard example admitting more than one solution.
What does the Picard iteration actually do?: It rewrites the initial value problem as an integral equation and repeatedly substitutes an approximate solution into the integral. When the right-hand side is Lipschitz, this iteration is a contraction, so it converges to the unique fixed point, which is the sought solution.