Distributions, also known as Schwartz distributions or generalized functions, are objects that generalize the classical notion of functions in mathematical analysis. Distributions make it possible to differentiate functions whose derivatives do not exist in the classical sense. In particular, any locally integrable function has a distributional derivative. Distributions are widely used in the theory of partial differential equations, where it may be easier to establish the existence of distributional solutions than classical solutions, or appropriate classical solutions may not exist. Distributions are also important in physics and engineering where many problems naturally lead to differential equations whose solutions or initial conditions are distributions, such as the Dirac delta function.
A function is normally thought of as acting on the points in its domain by "sending" a point x in its domain to the point Distribution theory reinterprets functions as linear functionals acting on test functions. Test functions are usually infinitely differentiable real (or complex) valued functions with compact support. Many "standard functions" (meaning for example a function that is typically encountered in a Calculus course), say for instance a continuous map can be canonically reinterpreted as acting on test functions (instead of their usual interpretation as acting on points of their domain) via the action of "integration against a test function"; explicitly, this means that "acts on" a test function g by "sending" g to the number This new action of is thus a real-valued map whose domain is the space of test functions; this map turns out to have two additional properties[note 1] that make it into what is known as a distribution on . Distributions that arise from "standard functions" in this way are the prototypical examples of a distributions. But there are many distributions that do not arise in this way and these distributions are the "generalized functions." Examples include the Dirac delta function or some distributions that arise via integration of test functions against measures. However, by using various methods it is nevertheless still possible to reduce any distribution down to a simpler family of related distributions that do arise via such actions of integration.
In applications to physics and engineering, the space of test functions usually consists of smooth functions with compact support that are defined on some given non-empty open subset This space of test functions is denoted by or and a distribution on U is by definition a linear functional on that is continuous when given a topology called the canonical LF topology. The space of (all) distributions on U, which is usually denoted by (note the prime), is thus the continuous dual space of and it is these distributions that are the main focus of this article.
There are other possible choices for the space of test functions, which lead to other different spaces of distributions. If then the use of Schwartz functions[note 2] as test functions gives rise to a certain subspace of whose elements are called tempered distributions. These are important because they allow the Fourier transform to be extended from "standard functions" to tempered distributions. The set of tempered distributions forms a vector subspace of the space of distributions and is thus one example of a space of distributions; there are many other spaces of distributions.
There also exist other major classes of test functions that are not subsets of , such as spaces of analytic test functions, which produce very different classes of distributions. The theory of such distributions has a different character from the previous one because there are no analytic functions with non-empty compact support.[note 3] Use of analytic test functions lead to Sato's theory of hyperfunctions.
The practical use of distributions can be traced back to the use of Green functions in the 1830s to solve ordinary differential equations, but was not formalized until much later. According to Kolmogorov & Fomin (1957) harvtxt error: no target: CITEREFKolmogorovFomin1957 (help), generalized functions originated in the work of Sergei Sobolev (1936) on second-order hyperbolic partial differential equations, and the ideas were developed in somewhat extended form by Laurent Schwartz in the late 1940s. According to his autobiography, Schwartz introduced the term "distribution" by analogy with a distribution of electrical charge, possibly including not only point charges but also dipoles and so on. Gårding (1997) comments that although the ideas in the transformative book by Schwartz (1951) were not entirely new, it was Schwartz's broad attack and conviction that distributions would be useful almost everywhere in analysis that made the difference.
Before proceeding we introduce some notation to simplify the expressions and statements below:
- Let be a function and use to denote its domain. The support of denoted by is the closure of the set in
- Throughout this article is a fixed positive integer and is a fixed non-empty open subset of
- For two functions set:
- A multi-index of size is simply an element in (given that is fixed we generally omit specifying the size of multi-indices). The length of a multi-index is defined as and denoted by Multi-indices are particularly useful when dealing with functions of several variables, in particular we introduce the following notations for a given multi-index :
- We also introduce a partial order of all multi-indices by if and only if for all When we define their multi-index binomial coefficient as:
Definitions of test functions and distributions
In this section, we will formally define real-valued distributions on U. With minor modifications, one can also define complex-valued distributions, and one can replace with any (paracompact) smooth manifold.
Notation: Suppose .
- Let denote the vector space of all k-times continuously differentiable real-valued functions on U.
- For any compact subset K ⊆ U, let and both denote the vector space of all those functions such that
- Note that depends on both K and U but we will only indicate K, where in particular, if then the domain of is U rather than K. We will use the notation } when only when the notation risks being ambiguous.
- Clearly, every contains the constant 0 map, even if K = ∅.
- Let denote the set of all such that for some compact subset K of U.
- Equivalently, is the set of all such that has compact support.
- is equal to the union of all as K ranges over 𝕂.
- Note that if is a real-valued function on U, then is an element of if and only if is a bump function.
Note that for all and any compact subsets K and L of U, we have:
Definition: Elements of are called test functions on U and is referred to as the space of test function. We will use both and to denote this space.
Distributions on U are defined to be the continuous linear functionals on the space of test functions on U, when it is endowed with a particular topology called the canonical LF topology. So to define the space of distributions we need a suitable topology on , which in turn requires that several other topological vector spaces (TVSs) be defined first. These are the spaces and for arbitrary
- Choice of compact sets 𝕂
Throughout, 𝕂 will be any collection of compact subsets of U such that (1) , and (2) for any compact K ⊆ U there exists some K2 ∈ 𝕂 such that K ⊆ K2. The most common choices for 𝕂 are:
- The set of all compact subsets of U, or
- A set where , and for all i, and Ui is a relatively compact non-empty open subset of U (i.e. "relatively compact" means that the closure of Ui, in either U or , is compact).
We make 𝕂 into a directed set by defining K1 ≤ K2 if and only if K1 ⊆ K2. Note that although the definitions of the subsequently defined topologies explicitly reference 𝕂, in reality they do not depend on the choice of 𝕂; that is, if 𝕂1 and 𝕂2 are any two such collections of compact subsets of U, then the topologies defined on and by using 𝕂1 in place of 𝕂 are the same as those defined by using 𝕂2 in place of 𝕂.
Topology on Ck(U)
We now introduce the seminorms that will define the topology on .
Definition: Suppose and K is an arbitrary compact subset of U. Suppose i an integer such that 0 ≤ i ≤ k[note 4] and p is a multi-index with length |p| ≤ k. For K ≠ ∅ we define:
while for K = ∅ we define all the functions above to be the constant 0 map.
Each of the following families of seminorms generates the same locally convex topology on :
With this topology, becomes a locally convex (non-normable) Fréchet space and all of the seminorms defined above are continuous on this space. Under this topology, a net in converges to if and only if for every multi-index p with |p| < k + 1 and every K ∈ 𝕂, the net converges to uniformly on K. For any any bounded subset of is a relatively compact subset of  In particular, a subset of is bounded if and only if it is bounded in for all  The space is a Montel space if and only if k = ∞.
The topology on is the superior limit of the subspace topologies induced on by the TVSs as i ranges over the non-negative integers. A subset W of is open in this topology if and only if there exists such that W is open when is endowed with the subspace topology induced by
Topology on Ck(K)
As before, fix Recall that if is any compact subset of then: .
For any compact subset K ⊆ U, is a closed subspace of the Fréchet space and is thus also a Fréchet space. For all compact K, L ⊆ U with K ⊆ L, denote the natural inclusion by where note that this map is a linear embedding of TVSs whose range is closed in its codomain (said differently, the topology on is identical to the subspace topology it inherits from , and is a closed subspace of ). The interior of relative to is empty.
And when k = 2, is even a Hilbert space. The space is a distinguished Schwartz Montel space so if then it is not normable and thus not a Banach space (although like all other , it is a Fréchet space).
Trivial extensions and independence of Ck(K)'s topology from U
The definition of depends on U so we will let denote the topological space , which by definition is a topological subspace of . Suppose V is an open subset of containing . Given we define its trivial extension to V, as follows:
Let denote the map that sends a function in to its trivial extension on V. This map is a linear injection and for every compact subset K ⊆ U, we clearly have where is the topological subspace of consisting of maps with support contained in K (since K ⊆ U ⊆ V, K is a compact subset of V as well). It follows that . If I is restricted to then the following induced map is a homeomorphism (and thus a TVS-isomorphism):
and thus the next two maps (which like the previous map are defined by ) are topological embeddings:
(the topology on is the canonical LF topology, which is defined later). Using we identify with its image in . Since , through this identification can also be considered as a subset of . Importantly, the subspace topology inherits from (when it is viewed as a subset of ) is identical to the subspace topology that it inherits from (when is viewed instead as a subset of via the identification). Thus the topology on is independent of the open subset U of that contains K. This justifies our practice of using instead of
Topology on the spaces of test functions and distributions
Recall that denote all those functions in that have compact support in U, where note that is the union of all as K ranges over 𝕂. Moreover, for every k, is a dense subset of . The spacial case when k = ∞ gives us the space of test functions.
Definition: is called the space of test functions on U and it may also be denoted by .
Canonical LF topology
We now define the canonical LF topology as a direct limit. It is also possible to define this topology in terms of its neighborhoods of the origin, which is described afterwards.
For any two sets K and L, we declare that K ≤ L if and only if K ⊆ L, which in particular makes the collection 𝕂 of compact subsets of U into a directed set (we say that such a collection is directed by subset inclusion). For all compact K, L ⊆ U with K ⊆ L, there are natural inclusions
Recall from above that the map is a topological embedding. The collection of maps
forms a direct system in the category of locally convex topological vector spaces that is directed by 𝕂 (under subset inclusion). This system's direct limit (in the category of locally convex TVSs) is the pair where are the natural inclusions and where is now endowed with the (unique) strongest locally convex topology making all of the inclusion maps continuous.
Assumption: We will henceforth assume that is endowed with its canonical LF topology.
- Neighborhoods of the origin
For all K ∈ 𝕂, is a neighborhood of the origin in .
Note that any convex set satisfying this condition is necessarily absorbing in . Since the topology of any topological vector space is translation-invariant, any TVS-topology is completely determined by the set of neighborhood of the origin. This means that one could actually define the canonical LF topology by declaring that a convex balanced subset U is a neighborhood of the origin if and only if it satisfies condition CN.
- Topology defined via differential operators
A linear differential operator in U with smooth coefficients is a sum
where and all but finitely many of are identically 0. The integer is called the order of the differential operator If is a linear differential operator of order k then it induces a canonical linear map defined by , where we shall reuse notation and also denote this map by .
For any 1 ≤ k ≤ ∞, the canonical LF topology on is the weakest locally convex TVS topology making all linear differential operators in U of order < k + 1 into continuous maps from into .
- Canonical LF topology's independence from 𝕂
One benefit of defining the canonical LF topology as the direct limit of a direct system is that we may immediately use the universal property of direct limits. Another benefit is that we can use well-known results from category theory to deduce that the canonical LF topology is actually independent of the particular choice of the directed collection 𝕂 of compact sets. And by considering different collections 𝕂 (in particular, those 𝕂 mentioned at the beginning of this article), we may deduce different properties of this topology. In particular, we may deduce that the canonical LF topology makes into a Hausdorff locally convex strict LF-space (and also a strict LB-space if k ≠ ∞), which of course is the reason why this topology is called "the canonical LF topology" (see this footnote for more details).[note 6]
- Universal property
From the universal property of direct limits, we know that if is a linear map into a locally convex space Y (not necessarily Hausdorff), then u is continuous if and only if u is bounded if and only if for every K ∈ 𝕂, the restriction of u to is continuous (or bounded).
- Dependence of the canonical LF topology on U
Suppose V is an open subset of containing Let denote the map that sends a function in to its trivial extension on V (which was defined above). This map is a continuous linear map. If (and only if) U ≠ V then is not a dense subset of and is not a topological embedding. Consequently, if U ≠ V then the transpose of is neither one-to-one nor onto.
- Bounded subsets
A subset B of is bounded in if and only if there exists some K ∈ 𝕂 such that and B is a bounded subset of . Moreover, if K ⊆ U is compact and then S is bounded in if and only if it is bounded in . For any 0 ≤ k ≤ ∞, any bounded subset of (resp. ) is a relatively compact subset of (resp. ), where ∞ + 1 = ∞.
For all compact K ⊆ U, the interior of in is empty so that is of the first category in itself. It follows from Baire's theorem that is not metrizable and thus also not normable (see this footnote[note 7] for an explanation of how the non-metrizable space can be complete even thought it doesn't admit a metric). The fact that is a nuclear Montel space makes up for the non-metrizability of (see this footnote for a more detailed explanation).[note 8]
- Relationships between spaces
Using the universal property of direct limits and the fact that the natural inclusions are all topological embedding, one may show that all of the maps are also topological embeddings. Said differently, the topology on is identical to the subspace topology that it inherits from , where recall that 's topology was defined to be the subspace topology induced on it by . In particular, both and induces the same subspace topology on . However, this does not imply that the canonical LF topology on is equal to the subspace topology induced on by ; these two topologies on are in fact never equal to each other since the canonical LF topology is never metrizable while the subspace topology induced on it by is metrizable (since recall that is metrizable). The canonical LF topology on is actually strictly finer than the subspace topology that it inherits from (thus the natural inclusion is continuous but not a topological embedding).
Indeed, the canonical LF topology is so fine that if denotes some linear map that is a "natural inclusion" (such as , or , or other maps discussed below) then this map will typically be continuous, which as is shown below, is ultimately the reason why locally integrable functions, Radon measures, etc. all induce distributions (via the transpose of such a "natural inclusion"). Said differently, the reason why there are so many different ways of defining distributions from other spaces ultimately stems from how very fine the canonical LF topology is. Moreover, since distributions are just continuous linear functionals on , the fine nature of the canonical LF topology means that more linear functionals on end up being continuous ("more" means as compared to a coarser topology that we could have placed on such as for instance, the subspace topology induced by some , which although it would have made metrizable, it would have also resulted in fewer linear functionals on being continuous and thus there would have been fewer distributions; moreover, this particular coarser topology also has the disadvantage of not making into a complete TVS).
- Other properties
- The differentiation map is a surjective continuous linear operator.
- The bilinear multiplication map given by is not continuous; it is however, hypocontinuous.
As discussed earlier, continuous linear functionals on a are known as distributions on U. Thus the set of all distributions on U is the continuous dual space of , which when endowed with the strong dual topology is denoted by .
One interprets this notation as the distribution T acting on the test function to give a scalar, or symmetrically as the test function acting on the distribution T.
- Characterizations of distributions
Proposition. If T is a linear functional on then the following are equivalent:
- T is a distribution;
- (definition) T is continuous;
- T is continuous at the origin;
- T is uniformly continuous;
- T is a bounded operator;
- T is sequentially continuous; i.e. for every sequence in that converges to 
- T is sequentially continuous at the origin; i.e. for every sequence in that converges to
- The kernel of T is a closed subspace of
- The graph of T is a closed;
- For every compact subset there exist constants and such that for all 
- For every compact subset there exist constants and such that for all with support contained in 
- For any compact subset and any sequence in if converges uniformly to zero for all multi-indices p, then
- Any of the three statements immediately above (i.e. statements 10, 11, and 12) but with the additional requirement that compact set K belongs to 𝕂.
Since the canonical LF topology is locally convex there are many collections of (necessarily continuous) seminorms on that define the canonical LF topology of .
Proposition. If T is a linear functional on then the following are equivalent:
- is continuous;
- There exists a continuous seminorm g on such that
- There exists a constant C > 0, a collection of continuous seminorms, that defines the canonical LF topology of and a finite subset such that 
Topology on the space of distributions
Definition and notation: The space of distributions on U, denoted by , is the continuous dual space of endowed with the topology of uniform convergence on bounded subsets of . More succinctly, the space of distributions on U is .
The topology of uniform convergence on bounded subsets is also called the strong dual topology.[note 9] This topology is chosen because it is with this topology that becomes a nuclear Montel space and it is with this topology that the kernels theorem of Schwartz holds. No matter what dual topology is placed on ,[note 10] a sequence of distributions converges in this topology if and only if it converges pointwise (although this need not be true of a net). No matter which topology is chosen, will be a non-metrizable, locally convex topological vector space. The space is separable and has the strong Pytkeev property but it is neither a k-space nor a sequential space, which in particular implies that it is not metrizable and also that its topology can not be defined using only sequences.
- Topological vector space categories
The canonical LF topology makes into a complete distinguished strict LF-space (and a strict LB-space if and only if k ≠ ∞), which implies that is a meager subset of itself. Furthermore, , as well as its strong dual space, is a complete Hausdorff locally convex barrelled bornological Mackey space. The strong dual of is a Fréchet space if and only if k ≠ ∞ so in particular, the strong dual of , which is the space of distributions on U, is not metrizable (note that the weak-* topology on also isn't metrizable and moreover, it further lacks almost all of the nice properties that the strong dual topology gives ).
The three spaces , , and the Schwartz space , as well as the strong duals of each of these three spaces, are complete nuclear Montel bornological spaces, which implies that all six of these locally convex spaces are also paracompact reflexive barrelled Mackey spaces. The spaces and are both distinguished Fréchet spaces. Moreover, both and are Schwartz TVSs.
- Convergent sequences and their insufficiency to describe topologies
The strong dual spaces of and are sequential spaces but not Fréchet-Urysohn spaces. Moreover, neither the space of test functions nor its strong dual is a sequential space (not even an Ascoli space), which in particular implies that their topologies can not be defined entirely in terms of convergent sequences.
A sequence in converges in if and only if there exists some K ∈ 𝕂 such that contains this sequence and this sequence converges in ; equivalently, it converges if and only if the following two conditions hold:
- There is a compact set K ⊆ U containing the supports of all
- For each multi-index α, the sequence of partial derivatives tends uniformly to .
Neither the space nor its strong dual is a sequential space, and consequently, their topologies can not be defined entirely in terms of convergent sequences. For this reason, the above characterization of when a sequence converges is not enough to define the canonical LF topology on . The same can be said of the strong dual topology on .
- What sequences do characterize
Nevertheless, sequences do characterize many important properties, as we now discuss. It is known that in the dual space of any Montel space, a sequence converges in the strong dual topology if and only if it converges in the weak* topology, which in particular, is the reason why a sequence of distributions converges (in the strong dual topology) if and only if it converges pointwise (this leads many authors to use pointwise convergence to actually define the convergence of a sequence of distributions; this is fine for sequences but it does not extend to the convergence of nets of distributions since a net may converge pointwise but fail to convergence in the strong dual topology).
Sequences characterize continuity of linear maps valued in locally convex space. Suppose X is a locally convex bornological space (such as any of the six TVSs mentioned earlier). Then a linear map F : X → Y into a locally convex space Y that maps null sequences[note 11] in X to bounded subsets of Y[note 12] is necessarily continuous. So in particular, if a linear map F : X → Y into a locally convex space is sequentially continuous at the origin then it is continuous. However, this does not necessarily extend to non-linear maps and/or to maps valued in topological spaces that are not locally convex TVSs.
- Sequences of distributions
A sequence of distributions converges with respect to the weak-* topology on to a distribution T if and only if
for every test function . For example, if is the function
and Tm is the distribution corresponding to , then
as m → ∞, so Tm → δ in . Thus, for large m, the function can be regarded as an approximation of the Dirac delta distribution.
- Other properties
- The strong dual space of is TVS isomorphic to via the canonical TVS-isomorphism defined by sending to value at (i.e. to the linear functional on defined by sending to );
- On any bounded subset of , the weak and strong subspace topologies coincide; the same is true for ;
- Every weakly convergent sequence in is strongly convergent (although this does not extend to nets).
Localization of distributions
There is no way to define the value of a distribution in at a particular point of U. However, as is the case with functions, distributions on U restrict to give distributions on open subsets of U. Furthermore, distributions are locally determined in the sense that a distribution on all of U can be assembled from a distribution on an open cover of U satisfying some compatibility conditions on the overlaps. Such a structure is known as a sheaf.
Restrictions to an open subset
Let U and V be open subsets of with V ⊆ U. Let be the operator which extends by zero a given smooth function compactly supported in V to a smooth function compactly supported in the larger set U. The transpose of is called the restriction mapping and is denoted by
The map is a continuous injection where if V ⊆ U then it is not a topological embedding and its range is not dense in , which implies that this map's transpose is neither injective nor surjective and that the topology that transfers from onto its image is strictly finer than the subspace topology that induces on this same set. A distribution is said to be extendible to U if it belongs to the range of the transpose of and it is called extendible if it is extendable to .
For any distribution , the restriction ρVU(T) is a distribution in defined by:
Unless U = V, the restriction to V is neither injective nor surjective. Lack of surjectivity follows since distributions can blow up towards the boundary of V. For instance, if U = ℝ and V = (0, 2), then the distribution
is in but admits no extension to .
Gluing and distributions that vanish in a set
Let V be an open subset of U. is said to vanishes in V if for all such that we have . T vanishes in V if and only if the restriction of T to V is equal to 0, or equivalently, if and only if T lies in the kernel of the restriction map ρVU.
- Corollary. Let be a collection of open subsets of and let T = 0 if and only if for each , the restriction of T to is equal to 0.
- Corollary. The union of all open subsets of U in which a distribution T vanishes is an open subset of U in which T vanishes.
Support of a distribution
This last corollary implies that for every distribution T on U, there exists a unique largest subset V of U such that T vanishes in V (and doesn't vanish in any open subset of U that is not contained in V); the complement in U of this unique largest open subset is called the support of T. Thus
If is a locally integrable function on U and if is its associated distribution, then the support of is the smallest closed subset of U in the complement of which is almost everywhere equal to 0. If is continuous, then the support of is equal to the closure of the set of points in U at which doesn't vanish. The support of the distribution associated with the Dirac measure at a point is the set . If the support of a test function does not intersect the support of a distribution T then Tf = 0. A distribution T is 0 if and only if its support is empty. If is identically 1 on some open set containing the support of a distribution T then fT = T. If the support of a distribution T is compact then it has finite order and furthermore, there is a constant C and a non-negative integer N such that:
If T has compact support then it has a unique extension to a continuous linear functional on ; this functional can be defined by , where is any function that is identically 1 on an open set containing the support of T.
If and then and . Thus, distributions with support in a given subset form a vector subspace of ; such a subspace is weakly closed in if and only if A is closed in U. Furthermore, if is a differential operator in U, then for all distributions T on U and all we have and 
Distributions with compact support
- Support in a point set and Dirac measures
For any , let denote the distribution induced by the Dirac measure at x. For any and distribution , the support of T is contained in if and only if T is a finite linear combination of derivatives of the Dirac measure at  If in addition the order of T is then there exist constants such that:
Said differently, if T has support at a single point then T is in fact a finite linear combination of distributional derivatives of the δ function at P. That is, there exists an integer m and complex constants aα such that
where is the translation operator.
- Distribution with compact support
- Distributions of finite order with support in an open subset
Global structure of distributions
The formal definition of distributions exhibits them as a subspace of a very large space, namely the topological dual of (or the Schwartz space for tempered distributions). It is not immediately clear from the definition how exotic a distribution might be. To answer this question, it is instructive to see distributions built up from a smaller space, namely the space of continuous functions. Roughly, any distribution is locally a (multiple) derivative of a continuous function. A precise version of this result, given below, holds for distributions of compact support, tempered distributions, and general distributions. Generally speaking, no proper subset of the space of distributions contains all continuous functions and is closed under differentiation. This says that distributions are not particularly exotic objects; they are only as complicated as necessary.
- Distributions as sheafs
Decomposition of distributions as sums of derivatives of continuous functions
By combining the above results, one may express any distribution on U as the sum of a series of distributions with compact support, where each of these distributions can in turn be written as a finite sum of distributional derivatives of continuous functions on U. In other words for arbitrary we can write:
where are finite sets of multi-indices and the functions are continuous.
Note that the infinite sum above is well-defined as a distribution. The value of T for a given can be computed using the finitely many gα that intersect the support of .
Operations on distributions
Many operations which are defined on smooth functions with compact support can also be defined for distributions. In general, if is a linear map which is continuous with respect to the weak topology, then it is possible to extend A to a map by passing to the limit.[note 13][clarification needed]
Preliminaries: Transpose of a linear operator
Operations on distributions and spaces of distributions are often defined by means of the transpose of a linear operator because it provides a unified approach that the many definitions in the theory of distributions and because of its many well-known topological properties. In general the transpose of a continuous linear map is the linear map defined by , or equivalently, it is the unique map satisfying for all and all . Since A is continuous, the transpose is also continuous when both duals are endowed with their respective strong dual topologies; it is also continuous when both duals are endowed with their respective weak* topologies (see the articles polar topology and dual system for more details).
In the context of distributions, the characterization of the transpose can be refined slightly. Let be a continuous linear map. Then by definition, the transpose of A is the unique linear operator that satisfies:
- for all and all .
However, since the image of is dense in , it is sufficient that the above equality hold for all distributions of the form where . Explicitly, this means that the above condition holds if and only if the condition below holds:
- for all .
Differentiation of distributions
Let is the partial derivative operator In order to extend we compute its transpose:
Therefore Therefore the partial derivative of with respect to the coordinate is defined by the formula
With this definition, every distribution is infinitely differentiable, and the derivative in the direction is a linear operator on .
More generally, if is an arbitrary multi-index, then the partial derivative of the distribution is defined by
Differentiation of distributions is a continuous operator on this is an important and desirable property that is not shared by most other notions of differentiation.
If T is a distribution in ℝ then
where is the derivative of T and τx is translation by x; thus the derivative of T may be viewed as a limit of quotients.
Differential operators acting on smooth functions
A linear differential operator in U with smooth coefficients acts on the space of smooth functions on Given we would like to define a continuous linear map, that extends the action of on to distributions on . In other words we would like to define such that the following diagram commutes:
Where the vertical maps are given by assigning its canonical distribution , which is defined by: for all . With this notation the diagram commuting is equivalent to:
In order to find we consider the transpose of the continuous induced map defined by As discussed above we may calculate the transpose using:
We now define the formal transpose of , which we'll denote by to avoid confusion with the transpose map, to be the following differential operator on U:
The computations above have shown that:
- Lemma. Let be a linear differential operator with smooth coefficients in Then for all we have
- which is equivalent to:
The Lemma combined with the fact that the formal transpose of the formal transpose is the original differential operator, i.e.  enables us to arrive at the correct definition: the formal transpose induces the (continuous) canonical linear operator defined by . We claim that the transpose of this map, , can be taken as . To see this we compute how it acts on a distribution of the form with :
We call the continuous linear operator the differential operator on distributions extending P. Its action on an arbitrary distribution is defined via:
If converges to then for every multi-index converges to
Multiplication of distributions by smooth functions
A differential operator of order 0 is just multiplication by a smooth function. And conversely, if is a smooth function then is a differential operator of order 0, whose formal transpose is itself (i.e. ). The induced differential operator maps a distribution T to a distribution denoted by . We have thus defined the multiplication of a distribution by a smooth function.
We now give an alternative presentation of multiplication by a smooth function. If is a smooth function and T is a distribution on U, then the product mT is defined by
This definition coincides with the transpose definition since if is the operator of multiplication by the function m (i.e., ), then
Under multiplication by smooth functions, is a module over the ring With this definition of multiplication by a smooth function, the ordinary product rule of calculus remains valid. However, a number of unusual identities also arise. For example, if δ′ is the Dirac delta distribution on ℝ, then mδ = m(0)δ, and if δ′ is the derivative of the delta distribution, then
Example. For any distribution T, the product of T with the function that is identically 1 on U is equal to T.
Example. Suppose is a sequence of test functions on U that converges to the constant function For any distribution T on U, the sequence converges to 
If converges to and converges to then converges to
Problem of multiplying distributions
It is easy to define the product of a distribution with a smooth function, or more generally the product of two distributions whose singular supports are disjoint. With more effort it is possible to define a well-behaved product of several distributions provided their wave front sets at each point are compatible. A limitation of the theory of distributions (and hyperfunctions) is that there is no associative product of two distributions extending the product of a distribution by a smooth function, as has been proved by Laurent Schwartz in the 1950s. For example, if p.v. 1/ is the distribution obtained by the Cauchy principal value
If δ is the Dirac delta distribution then
so the product of a distribution by a smooth function (which is always well defined) cannot be extended to an associative product on the space of distributions.
Thus, nonlinear problems cannot be posed in general and thus not solved within distribution theory alone. In the context of quantum field theory, however, solutions can be found. In more than two spacetime dimensions the problem is related to the regularization of divergences. Here Henri Epstein and Vladimir Glaser developed the mathematically rigorous (but extremely technical) causal perturbation theory. This does not solve the problem in other situations. Many other interesting theories are non linear, like for example the Navier–Stokes equations of fluid dynamics.
Several not entirely satisfactory theories of algebras of generalized functions have been developed, among which Colombeau's (simplified) algebra is maybe the most popular in use today.
Inspired by Lyons' rough path theory, Martin Hairer proposed a consistent way of multiplying distributions with certain structure (regularity structures), available in many examples from stochastic analysis, notably stochastic partial differential equations. See also Gubinelli–Imkeller–Perkowski (2015) for a related development based on Bony's paraproduct from Fourier analysis.
Composition with a smooth function
Let T be a distribution on Let V be an open set in , and F : V → U. If F is a submersion, it is possible to define
This is the composition of the distribution T with F, and is also called the pullback of T along F, sometimes written
The pullback is often denoted F*, although this notation should not be confused with the use of '*' to denote the adjoint of a linear mapping.
The condition that F be a submersion is equivalent to the requirement that the Jacobian derivative dF(x) of F is a surjective linear map for every x ∈ V. A necessary (but not sufficient) condition for extending F# to distributions is that F be an open mapping. The inverse function theorem ensures that a submersion satisfies this condition.
If F is a submersion, then F# is defined on distributions by finding the transpose map. Uniqueness of this extension is guaranteed since F# is a continuous linear operator on . Existence, however, requires using the change of variables formula, the inverse function theorem (locally) and a partition of unity argument.
In the special case when F is a diffeomorphism from an open subset V of onto an open subset U of change of variables under the integral gives
In this particular case, then, F# is defined by the transpose formula:
Under some circumstances, it is possible to define the convolution of a function with a distribution, or even the convolution of two distributions. Recall that if and g are functions on then we denote by the convolution of and g, defined at to be the integral
provided that the integral exists. If are such that 1/r = (1/p) + (1/q) - 1 then for any functions and we have and . If and g are continuous functions on , at least one of which has compact support, then and if then the value of on A do not depend on the values of outside of the Minkowski sum 
Importantly, if has compact support then for any , the convolution map is continuous when considered as the map or as the map .
- Translation and symmetry
Given , the translation operator τa sends to defined by This can be extended by the transpose to distributions in the following way: given a distribution T, the translation of T by a is the distribution defined by 
Given define the function by . Given a distribution T, let be the distribution defined by . The operator is called the symmetry with respect to the origin.
Convolution of a test function with a distribution
Convolution with defines a linear map:
Convolution of with a distribution can be defined by taking the transpose of Cf relative to the duality pairing of with the space of distributions (Trèves 1967, Chapter 27) harv error: no target: CITEREFTrèves1967 (help). If , then by Fubini's theorem
Extending by continuity, the convolution of with a distribution T is defined by
An alternative way to define the convolution of a test function and a distribution T is to use the translation operator τa. The convolution of the compactly supported function and the distribution T is then the function defined for each by
It can be shown that the convolution of a smooth, compactly supported function and a distribution is a smooth function. If the distribution T has compact support then if is a polynomial (resp. an exponential function, an analytic function, the restriction of an entire analytic function on to , the restriction of an entire function of exponential type in to ) then the same is true of . If the distribution T has compact support as well, then is a compactly supported function, and the Titchmarsh convolution theorem (Hörmander 1983, Theorem 4.3.3) implies that
where ch denotes the convex hull and supp denotes the support.
Convolution of a smooth function with a distribution
Let and and assume that at least one of and T has compact support. The convolution of and T, denoted by or by , is the smooth function:
If T is a distribution then the map is continuous as a map where if in addition T has compact support then it is also continuous as the map and continuous as the map .
If is a continuous linear map such that for all and all then there exists a distribution such that for all .
Let be the Dirac measure at 0 and its derivative as a distribution. Then and . Importantly, the associative law fails to hold:
Convolution of distributions
It is also possible to define the convolution of two distributions S and T on , provided one of them has compact support. Informally, in order to define S ∗ T where T has compact support, the idea is to extend the definition of the convolution ∗ to a linear operation on distributions so that the associativity formula
continues to hold for all test functions .
It is also possible to provide a more explicit characterization of the convolution of distributions (Trèves 1967, Chapter 27) harv error: no target: CITEREFTrèves1967 (help). Suppose that S and T are distributions and that S has compact support. Then the linear maps
are continuous. The transposes of these maps,
are consequently continuous and one may show that
This common value is called the convolution of S and T and it is a distribution that is denoted by or . It satisfies . If S and T are two distributions, at least one of which has compact support, then for any , . If T is a distribution in and if is a Dirac measure then .
Suppose that it is T that has compact support. For consider the function
It can be readily shown that this defines a smooth function of x, which moreover has compact support. The convolution of S and T is defined by
This generalizes the classical notion of convolution of functions and is compatible with differentiation in the following sense: for every multi-index α,
This definition of convolution remains valid under less restrictive assumptions about S and T.
The convolution of distributions with compact support induces a continuous bilinear map