Distribution (mathematical analysis)

Distributions, also known as generalized functions, extend the idea of regular functions in mathematical analysis. They allow us to work with things that aren’t usually considered functions, helping us find answers to problems that would otherwise be impossible. For example, distributions let us differentiate functions that don’t have derivatives in the normal way.

One famous distribution is the Dirac delta function, which is important in physics and engineering. It acts like an infinitely sharp spike that is zero everywhere except at one point, where its total area is one. This idea helps solve many real-world problems, especially those described by partial differential equations.

The concept of distributions began with efforts to solve equations using Green functions in the 1800s. It was formally developed in the mid-20th century by mathematicians like Sergei Sobolev and Laurent Schwartz. Their work created tools that are now used in many areas of science and mathematics.

Basic idea

Distributions are special tools in math that help us work with functions in new ways. They let us find "derivatives" even when normal rules don't work. For example, the Dirac delta is a famous distribution that picks out a single point — like finding the value of a function exactly at zero.

We can think of distributions as rules that tell us how to combine with special, well-behaved functions. These rules let us do calculus in more ways than before, helping us solve harder problems.

Test functions and distributions

Main article: Spaces of test functions and distributions

Distributions, also known as generalized functions, extend the idea of regular functions in mathematical analysis. They allow us to differentiate functions even when traditional derivatives do not exist. This is especially useful because any function that can be integrated locally has a distributional derivative.

Test functions are smooth functions with compact support, meaning they are non-zero only within a limited region. These functions help define distributions, which are continuous linear functionals that act on test functions. Essentially, a distribution assigns a real or complex number to each test function in a way that respects linear combinations and continuity. One key example is how a sequence of functions can approximate the Dirac delta distribution, which is important in physics and engineering.

Operations on distributions

Many operations that work on smooth functions can also be used on distributions. These operations help us understand and work with functions that don’t have traditional derivatives in the usual way.

One key idea is that we can extend these operations to distributions using a method called the "transpose." This means we use a special rule to define how these operations act on distributions. For example, when we take the derivative of a distribution, we use integration by parts, which lets us define the derivative even when the original function isn’t smooth. This way, every distribution can be differentiated many times, which is very useful in advanced mathematics.

Localization of distributions

Distributions are a way to generalize functions in mathematics. Unlike regular functions, distributions don’t have a value at a single point, but they can be studied over areas or regions.

When we look at a smaller part of a bigger area, we can “restrict” a distribution to that smaller area. This helps us understand how the distribution behaves in different places. We can also describe where a distribution is “active” by looking at its support — the region where it doesn’t vanish (or become zero). Some distributions have compact support, meaning their activity is limited to a small, closed area. This idea is important in advanced studies of functions and calculus.

Main article: Sheaf

Tempered distributions and Fourier transform

"Tempered distribution" redirects here. For tempered distributions on semisimple groups, see Tempered representation.

Tempered distributions are special kinds of generalized functions. They help us study the Fourier transform, which is a way to change functions into other forms that can be easier to work with. All tempered distributions have a Fourier transform, unlike some other generalized functions.

These distributions use a specific set of test functions called the Schwartz space. These functions change very smoothly and quickly become very small as their input grows. Because of this, the Fourier transform of a Schwartz function is also a Schwartz function. This makes tempered distributions very useful in studying Fourier transforms.

Convolution

Sometimes, we can combine a regular function with a special kind of object called a distribution, or even combine two of these objects, through a process called convolution. This helps us work with functions that are not easy to handle normally.

When we convolve a smooth, compactly supported function with a distribution, the result is always a smooth function. If the distribution also has compact support, the result will also have compact support. This idea extends the usual way we think about combining functions and works well with taking derivatives.

Distributions as derivatives of continuous functions

Distributions are special tools in math that help us understand things even when regular rules don’t work. They let us find "derivatives" of functions that don’t usually have them. Think of it like finding patterns where things change, even if those changes aren’t smooth or clear at first glance.

Basically, any distribution can be thought of as a derivative of a continuous function. This means distributions aren’t as strange as they might seem — they’re just built from everyday functions in a clever way. This idea helps mathematicians study all sorts of complex patterns and changes.

Main article: Schwartz functions
Main articles: Multi-indices, Rudin 1991

Using holomorphic functions as test functions

The idea of using special types of functions called holomorphic functions as a foundation for more advanced math led to new discoveries. Researchers, including Mikio Sato, developed a refined theory using tools like sheaf theory and several complex variables. This work helped make certain tricky math ideas, such as Feynman integrals, more precise and reliable.

Problem of multiplication

Multiplying distributions can be tricky. We can easily multiply a distribution by a smooth function, or two distributions if their "problem spots" don't overlap. However, there is a big limitation: we cannot always multiply two distributions together in a way that follows the usual rules of math.

For example, scientists have shown that some ways of multiplying distributions give different results depending on the order you do the multiplication. This means we can't always solve certain complex math problems using just distributions.

But there are special methods, like those used in quantum field theory, that can help solve some of these problems. Scientists continue to develop new theories to better understand how to multiply distributions in different situations.