nLab multi-valued function (changes)

Showing changes from revision #9 to #10: Added | ~~Removed~~ | ~~Chan~~ged

Multi-valued functions

Idea
Definitions
- With power sets
- Without power sets
Examples
Related entries
References

Idea

A multi-valued function $f: A \to B$ is like a function from $A$ to $B$ except that there may be more than one possible value $f(x)$ for a given element $x$ of $A$ . (Compare a partial function, where $f(x)$ may not exist at all.)

In older literature (into the 20th century, especially in analysis), functions were often considered to be multi-valued by default, requiring one to specify a singe-valued function otherwise. As set-theoretic formalisation spread, this intuition became difficult to maintain, and the modern concept of function must be single-valued. If you want multi-valued functions, then you can get them in terms of single-valued functions as below.

Definitions

~~Given sets $A$ and $B$ , a multi-valued function $f$ from a $A$ to $B$ is a span~~

With power sets

~~$\array { & & D \\ & \swarrow_\pi & & \searrow^f \\ A & & & & B \\ }$~~

Given sets $A$ and $B$ , a multi-valued function $f$ from $A$ to $B$ is a function $f$ from $A$ to the power set $\mathcal{P}(B)$ such that for each element $x$ in $A$ the subset $f(x)$ of $B$ is inhabited. By uncurrying the function one gets an entire relation.

~~of single-valued functions, where $\pi: D \to A$ is a surjection. (This condition can be dropped to define a multi-valued partial function, which is simply a span.)~~

Without power sets

We Given ~~will~~ ~~call~~ ~~$A$~~ set s ~~and~~ $B A$ B A ~~the~~ and~~source~~ $B$ , ~~and~~ a~~target~~multi-valued function of $f$ as from ~~usual;~~ a ~~then~~ we ~~call~~ $D A$ D A ~~the~~ to~~domain~~ $B$ of is a ~~$f$~~ span ~~and~~ ~~$\pi: D \to A$~~ ~~the~~ ~~projection~~ ~~of the domain onto the source. By abuse of notation, the multi-valued function~~ ~~$f$~~ ~~is conflated with the (single-valued) function~~ ~~$f: D \to B$~~ .

Often one can assume that the induced function $D \to A \times B$ is an injection; in that case, a multi-valued function is the same as an entire relation. On the other hand, if you're considering all of the multi-valued functions for a given $D$ , then this restriction is not really appropriate.

\array { & & D \\ & \swarrow_\pi & & \searrow^f \\ A & & & & B \\ }

of single-valued functions, where $\pi: D \to A$ is a surjection. (This condition can be dropped to define a multi-valued partial function, which is simply a span.)

We will call $A$ and $B$ the source and target of $f$ as usual; then we call $D$ the domain of $f$ and $\pi: D \to A$ the projection of the domain onto the source. By abuse of notation, the multi-valued function $f$ is conflated with the (single-valued) function $f: D \to B$ .

We consider two multi-valued functions (with the same given source and target) to be equal if there is a bijection between their domains that makes the obvious diagrams commute.

Examples

In 19th-century analysis, one considered the square-root function, the logarithm, and so forth to be multi-valued functions of complex numbers. We now understand this in terms of Riemann surfaces; the domain $D$ above is a Riemann surface. (Notice that the logarithm is actually a multi-valued partial function from $\mathbf{C}$ to $\mathbf{C}$ , although it is a multi-valued total function on $\mathbf{C} \setminus \{0\}$ .)
On anyon-wavefunctions as multi-valued functions on a configuration space of points: see there