Adjoint functors

In the field of category theory, an adjunction describes a specific relationship between two functors.

This relationship acts as a weak form of equivalence between two related categories. When two functors exhibit this connection, they are known as adjoint functors. One functor is designated as the left adjoint, while the other is the right adjoint. This pairing is ubiquitous throughout various branches of mathematics. These functors often arise when mathematicians seek optimal solutions to specific problems. These solutions are characterized by a concept known as a universal property.

An adjunction can be understood as a formulaic method for finding the most efficient solution to a problem. To be formulaic means the construction follows a consistent rule that works the same way every time. In many cases, a construction is considered most efficient if it satisfies a universal property. These properties generally fall into two types: initial properties and terminal properties. An initial property involves setting up a problem within an auxiliary category. The goal is then to find an initial object within that category. This process makes the optimization rigorous, much like finding a supremum in other mathematical contexts.

Consider the problem of turning a rng into a ring in ring theory. A rng is a structure similar to a ring but lacks a multiplicative identity. The most efficient way to solve this is to adjoin a single element, '1'. One must add only the elements necessary to satisfy the ring axioms. No extra relations should be imposed unless they are forced by those axioms. This specific construction defines a functor. In this scenario, the process of adjoining an identity is the left adjoint functor. The process of "forgetting" the identity to treat a ring as a rng is the right adjoint.

Mathematicians use several equivalent definitions to describe these adjoint relationships. The definition via universal morphisms is often the easiest to use when constructing an adjoint. A functor is a left adjoint if, for each object in the target category, there exists a universal morphism. This requires that for every other object and morphism, there exists a unique morphism that makes a specific diagram commute.

This definition is highly intuitive because it mirrors the logic of optimization problems. Similarly, a right adjoint can be defined using a universal morphism from an object to a target. These definitions are minimal in their requirements, making them very useful for proofs.

A second way to define an adjunction is through the use of hom-sets. This definition uses a natural isomorphism between two different sets of morphisms.

This approach highlights the inherent symmetry of the relationship. It is the reason why the term "adjoint" is used. For locally small categories, this naturality implies specific isomorphisms between pairs of functors. While this definition is a logical compromise, it serves as a vital stepping stone between other methods. It allows mathematicians to see the balance between the left and right sides of the relationship.

A third method involves the use of the unit and the counit. These are natural transformations that facilitate movement between the two categories. The compositions of these transformations must result in identity morphisms. This is often expressed through the counit–unit equations. This definition is particularly convenient for performing algebraic manipulations during proofs. Because all these definitions are equivalent, mathematicians can switch between them as needed. This flexibility allows them to avoid repeating complex details in different subject areas.

Adjoint functors are deeply connected to many other mathematical structures and fields. They are closely related to the concept of adjoint operators in Hilbert spaces. The terminology itself is derived from this mathematical analogy. General theorems about adjoint functors provide deep insights into many mathematical results. For example, left adjoints preserve colimits, while right adjoints preserve limits. These concepts are foundational to almost every area of mathematics.

By studying these pairs, researchers can uncover the underlying connections between algebra, topology, and other complex systems.