homotopy theory, (∞,1)-category theory, homotopy type theory
flavors: stable, equivariant, rational, p-adic, proper, geometric, cohesive, directed…
models: topological, simplicial, localic, …
see also algebraic topology
Introductions
Definitions
Paths and cylinders
Homotopy groups
Basic facts
Theorems
model category, model -category
Definitions
Morphisms
Universal constructions
Refinements
Producing new model structures
Presentation of -categories
Model structures
for -groupoids
on chain complexes/model structure on cosimplicial abelian groups
related by the Dold-Kan correspondence
for equivariant -groupoids
for rational -groupoids
for rational equivariant -groupoids
for -groupoids
for -groups
for -algebras
general -algebras
specific -algebras
for stable/spectrum objects
for -categories
for stable -categories
for -operads
for -categories
for -sheaves / -stacks
The Bousfield-Kan formula (BK72) is a presentation of colimits (or dually, limits) in -categories (i.e., homotopy colimits) in terms of coproducts and colimits of simplicial objects.
In ordinary category theory, the colimit of a diagram may be computed as a coequalizer of the parallel pair
When is an -category, we need to take “higher data” into account. So we take the geometric realization (i.e., -indexed colimit) of the simplicial object
This can be formalized as follows: There is an “initial vertex” functor from the opposite of the category of simplices of , which is homotopy final. Thus,
The simplicial object is just the left Kan extension of along the projection , so we have
In a simplicial model category , homotopy colimits indexed by are modeled by geometric realization (see there). Therefore, the Bousfield-Kan formula for the homotopy colimit of a (pointwise cofibrant) diagram is computed by
Textbook account includes Rie14, Chapter 5.
Let be a (pointwise cofibrant) diagram in a simplicial model category. Some coend calculus shows Rie14, 6.6.1 that
The right hand side is also called the Bousfield-Kan formula.
A cosimplicial framing on a model category is roughly a functorial choice of cosimplicial resolutions Hir03, Chapter 16. With this data, we can formally mimic the Bousfield-Kan formula, and this is adopted as the definition of homotopy colimits in Hir03, Chapter19; that this approach is equivalent to the ordinary notion of homotopy colimits is established in AO23 (for combinatorial model categories) and Ara23, Theorem 1.16 (for general model categories).
In a simplicial model category (this can be generalized, as discussed above), there are two formulas for homotopy colimits of simplicial objects: The Bousfield–Kan formula, and geometric realization. The Bousfield–Kan map(s) compare these formulas.
More precisely, let be an (SSet,)-enriched category and write
for the canonical cosimplicial simplicial set (the adjunct of the hom-functor ).
Write furthermore for the fat simplex, the cosimplicial simplicial set which assigns to the nerve of the overcategory .
The Bousfield–Kan map of cosimplicial simplicial maps is a canonical morphism
of cosimplicial simplicial sets.
This can also be regarded as a morphism
This morphism induces the following morphisms between (co)simplicial objects in .
For any simplicial object in , the realization of is the coend
where in the integrand we have the copower (or tensoring) of by sSet.
Here the Bousfield–Kan map is the morphism
There is also the Bousfield–Kan map for cosimplicial objects:
For any cosimplicial object, its totalization is the -weighted limit
where in the integrand we have the power or cotensor of by SSet.
Here the Bousfield–Kan morphism is the morphism
If the simplicial object is Reedy-cofibrant then its Bousfield–Kan map is a natural weak equivalence.
If the co-simplicial object is Reedy fibrant then its Bousfield–Kan map is a natural weak equivalence.
This can be proven for instance using homotopy colimits in the Reedy model structure. Details are at Reedy model structure – over the simplex category.
When the cosimplicial object is degreewise fibrant, then
computes the homotopy limit of as a weighted limit (as explained there). Then the above theorem says that the homotopy limit is already computed by the totalization
The Bousfield-Kan formula was introduced in:
Discussion in model categories:
Philip Hirschhorn, Model Categories and Their Localizations, AMS Math. Survey and Monographs 99 (2002) [ISBN:978-0-8218-4917-0, pdf toc, pdf, pdf]
The Bousfield–Kan map(s) are on p. 397, def. 18.7.1 and def. 18.7.3.
Realization and totalization are defs 18.6.2 and 18.6.3 on p. 395.
Beware that this book writes for the nerve!
Discussions in simplicial model categories:
See also:
Sergey Arkhipov, Sebastian Ørsted: Homotopy (co)limits via homotopy (co)ends in general combinatorial model categories, Appl. Categ. Structures 31 6 (2023) [doi:10.1007/s10485-023-09747-8]
Kensuke Arakawa, Section 1 of: Homotopy Limits and Homotopy Colimits of Chain Complexes, (2023) [arXiv:2310.00201]
Last revised on February 16, 2025 at 10:36:07. See the history of this page for a list of all contributions to it.