Foundations of category theory: categories, functors, natural transformations, limits, adjunctions, and abelian categories.
Core idea
This section covers the foundations of category theory, providing a unified language for describing mathematical structures and the relationships between them.
Let C be a category. An object of C is an element of the collection Ob(C).
An object has no “internal data” in the definition beyond belonging to C; its properties are typically expressed by the pattern of morphisms to and from it.
Equivalently, η is a natural isomorphism iff there exists a natural transformation η−1:G⇒F such that ηX−1=(ηX)−1 for all X (and hence η−1∘η=idF, η∘η−1=idG).
A product is unique up to unique isomorphism: if (P,π1,π2) and (P′,π1′,π2′) are both products of A,B, there is a unique isomorphism P≅P′ compatible with projections.
This is a special case of a limit (the limit of the discrete diagram AB).
Kernels are a categorical version of “solutions of f(x)=0”, defined using universal properties.
Throughout, assume C is a category with a zero object (e.g. any additive category), so that for any objects A,B there is a distinguished zero morphism0A,B:A→B.
An abelian category is most often defined as an additive category in which kernels/cokernels exist and monomorphisms/epimorphisms are “normal.” One standard axiom list is:
These notes are a scratch space for category-theory diagrams. The mathematical content is intentionally compact; the main point is to exercise the renderer on commutative diagrams, universal properties, adjunction triangles, and string-like composition pictures.